0% found this document useful (0 votes)
60 views180 pages

Module 18 IoT and OT

Chapter 18 discusses the integration of the Internet of Things (IoT) and Operational Technology (OT), highlighting the challenges and security risks associated with IoT devices across various industries. It outlines the fundamental concepts of IoT, the architecture, application areas, and the various technologies and protocols that facilitate communication within IoT networks. The chapter also emphasizes the importance of implementing protective measures against potential threats and vulnerabilities in IoT and OT systems.

Uploaded by

Vaino Uusiku
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
60 views180 pages

Module 18 IoT and OT

Chapter 18 discusses the integration of the Internet of Things (IoT) and Operational Technology (OT), highlighting the challenges and security risks associated with IoT devices across various industries. It outlines the fundamental concepts of IoT, the architecture, application areas, and the various technologies and protocols that facilitate communication within IoT networks. The chapter also emphasizes the importance of implementing protective measures against potential threats and vulnerabilities in IoT and OT systems.

Uploaded by

Vaino Uusiku
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Chapter 18: IoT and OT Hacking

Introduction
The Internet of Things (IoT) has emerged by integrating wireless technologies, micro-
electromechanical systems, microservices, and the internet. IoT applications span various
industries, including healthcare, agriculture, building management, transportation, and energy.
Numerous organizations are spearheading the IoT revolution. Smart wearables, industrial
machinery, connected electronics, smart grids, and vehicles are becoming integral to
interconnected networks. These devices generate vast amounts of data gathered, analyzed,
stored, and managed within these networks.
IoT has brought new technologies and capabilities into everyday life. However, as an evolving
domain, its technological and service immaturity from different vendors poses challenges for
organizations, especially security. Ensuring IoT security is complicated due to devices using
simple processors and basic operating systems, which often lack support for advanced security
measures. Businesses leveraging IoT must safeguard devices and data from potential threats.
Cybersecurity has become critical as industries increasingly digitize operations to improve
efficiency through remote data access and internet connectivity. The convergence of
Operational Technology (OT) and Information Technology (IT) introduces unique risks and
safety concerns. To address these challenges, organizations must thoroughly assess their cyber
threat landscape, industrial systems, and business processes. Prioritizing and addressing the
most significant risks and vulnerabilities is essential before implementing cybersecurity
measures and policies.
This chapter identifies potential risks associated with IoT and OT platforms while offering
strategies to safeguard IoT devices and OT systems against evolving threats and vulnerabilities.
By the end of this chapter, you will be able to:
● Understand IoT fundamentals
● Recognize various IoT threats and attack vectors
● Outline the methodology used in IoT hacking
● Utilize IoT hacking tools effectively
● Implement protective measures to defend against IoT-based attacks
● Operate IoT security tools
● Grasp the core concepts of OT systems
● Identify threats and attack techniques targeting OT environments
● Detail the process of OT hacking and the tools used in such scenarios
● Employ countermeasures to secure industrial operations against OT-related threats
● Leverage OT security tools for enhanced protection.

IoT Hacking
IoT hacking exploits flaws in Internet of Things (IoT) networks or devices to obtain illegal
access, interfere with normal operations, or steal information. These assaults target gadgets
such as wearables, industrial sensors, and smart home systems, frequently using weak security
measures to infiltrate the system.
IoT Concepts and Attacks
The Internet of Things (IoT) is a significant and developing subject in technology, economy,
and society. This convergence of big data analytics and machine-to-machine communications
is known as the "web of connected devices." The IoT is a forward-looking evolution of the
internet and physical device capabilities, progressively reducing the distance between the
virtual and real worlds. This part covers key IoT ideas necessary to comprehend the more
complex subjects discussed later in this chapter.
Attackers employ various tactics to target IoT devices or networks. This section covers the main
IoT dangers and the fundamental IoT attack vectors and tactics, such as Distributed Denial-of-
Service (DDoS) assaults, Heating, Ventilation, and Air Conditioning (HVAC) system attacks,
rolling code attacks, BlueBorne attacks, and jammer attacks.
What is the IoT?
The Internet of Things (IoT), also referred to as the Internet of Everything (IoE), describes a
network of web-connected devices equipped with sensors, processors, and communication
hardware. These devices can gather, process, and transmit data. In this context, a "thing"
represents any device embedded within natural, human-made, or machine-made objects that
can communicate over a network.
Key characteristics of IoT include connectivity, intelligent sensors, artificial intelligence,
compact design, and interactive engagement.
How does the IoT Work?
IoT technology comprises four main components: IoT devices, gateway systems, cloud-based
data storage, and remote control through mobile applications. These elements work together
to enable seamless communication between two endpoints. Key components of IoT technology
that ensure the efficient functioning of IoT devices include:
● Sensing Technology: Devices are equipped with sensors to detect and collect various types
of data from their environment, such as temperature, gas levels, location, machinery
operations, or patient health metrics.
● IoT Gateways: These act as intermediaries, linking IoT devices (internal networks) with
end-users (external networks). They facilitate data transfer from sensors to either the user
or the cloud.
● Cloud Storage and Data Processing: Once data passes through the gateway, it is stored
and analyzed on cloud servers. The processed information is then sent to the user for
decision-making or action.
● Remote Control via Mobile Apps: Users can manage, monitor, and control IoT devices
remotely using apps on smartphones, tablets, or computers. These apps enable data retrieval
and execution of specific actions from anywhere.
Example:
1. A home smart security system connects to the internet and cloud infrastructure via a
gateway, enabling seamless integration and communication.
2. The cloud stores detailed data about all devices linked to the network, including device IDs,
current status, access history, frequency of access, and duration of previous uses.
3. The connection between devices and the cloud server is established using web services.
4. A user with the appropriate mobile app can remotely interact with the device. To ensure
security, the user must first authenticate their identity. Access is granted if the credentials
match the records stored in the cloud; otherwise, it is denied. The cloud server identifies
the device by its ID and processes the corresponding request through the gateway.
5. Suppose the security system detects suspicious activity while recording at home. In that
case, it sends an alert to the cloud via the gateway. The cloud verifies the device ID and its
associated user, then forwards the alert to the end-user for prompt action.

Figure 18-01: Workings of the IoT

IoT Architecture
The IoT architecture is structured across multiple layers, from the top application layer to the
base edge technology layer. This design enables the system to cater to the needs of diverse
sectors such as communities, industries, enterprises, and governments. The roles of each layer
are outlined below:
● Edge Technology Layer: This foundational layer encompasses hardware components like
sensors, RFID tags, readers, and soft sensors, along with the devices themselves. These
elements are critical for collecting data and monitoring or sensing various phenomena. This
layer gathers data, connects devices within the network, and links them to the server.
● Access Gateway Layer: This layer facilitates initial data processing by serving as a bridge
between endpoints like devices and clients. Key functions include routing messages,
identifying them, and managing subscriptions.
● Internet Layer: A vital layer for enabling communication, it handles interactions such as
device-to-device, device-to-cloud, device-to-gateway, and back-end data exchanges.
● Middleware Layer: Positioned between the hardware and application layers, this layer acts
as an interface and operates in both directions. It manages data and devices, performing
tasks like data analysis, aggregation, filtering, access control, and device information
discovery.
● Application Layer: At the top of the architecture, this layer delivers services to users across
sectors such as construction, manufacturing, healthcare, automotive, and security, ensuring
tailored solutions for different needs.
• Delivery of various
Application applications to different
Layer users in IoT Devices

Middleware • Device management and


Layer information management

Internet • Connection between


Layer endpoints

Access • Protocol translation and


Gateway messaging
Layer

Edge • Sensors, devices, machines,


and intelligent edge nodes
Technology of various types
Layer

Figure 18-02: IoT Architecture

IoT Application Areas and Devices


IoT devices find applications across numerous fields, playing a vital role in simplifying daily
tasks, automating processes, and enhancing quality of life. IoT technology is deeply integrated
into various aspects of modern society, from smart homes to industrial systems. Here are some
key applications of IoT devices:
● Smart Home and Building Systems: Internet-connected devices, including thermostats,
lighting systems, and security solutions, provide services to enhance convenience and
efficiency in residential and commercial spaces.
● Healthcare and Life Sciences: IoT devices in this sector include wearable health trackers,
monitoring tools like implanted pacemakers, ECG and EKG machines, surgical equipment,
and telemedicine systems, enabling better healthcare delivery and patient outcomes.
● Industrial IoT (IIoT): IIoT is revolutionizing industries by increasing productivity,
integrating intelligent technology to transform manufacturing processes, and introducing
innovative hybrid business models.
● Transportation: IoT technology enables vehicle-to-vehicle, vehicle-to-roadside, and
vehicle-to-pedestrian communication, improving traffic flow, navigation systems, and
parking management.
● Retail: IoT is widely used for tasks like secure payment processing, targeted advertising,
and product monitoring to prevent theft and losses, thereby enhancing revenue generation.
● IT and Networking: Office IoT devices include printers, fax machines, copiers, and PBX
monitoring systems. These devices improve endpoint communication and facilitate efficient
data transmission over long distances.

Service Sectors Application Groups Locations Devices


Office, Education, Heating, Ventilation,
Commercial/ Retail, Hospitality, and Air
Institutional Healthcare, Airports, Conditioning
Buildings Stadiums (HVAC), Transport,
Fire and Safety,
Process, Clean Lighting, Security,
Industrial
Room, Campus Access, etc.
Power Generation,
Transport, and
Distribution, Low
Supply/Demand
Voltage, Power
Quality, Energy Turbines, Windmills,
Management UPS, Batteries,
Energy
Solar Wind, Co- Generators, Meters,
Alternative generation, Drills, Fuel Cells, etc.
Electrochemical
Rigs, Derricks,
Oil/Gas Heads, Pumps,
Pipelines
Wiring, Network Digital Cameras,
Infrastructure Access, Energy Power Systems, MID,
Management E-Readers,
Security/Alerts, Fire Dishwashers,
Awareness and Safety, Elderly, Desktop Computers,
Consumer and
Safety Children, Power Washing
Home
Protection Machines/Dryers,
Meters, Lights, TVs,
HVAC/Climate, MP3 devices, game
Convenience and
Lighting, Appliances, consoles, Alarms,
Entertainment
Entertainment etc.
Hospital, ER, Mobile,
Care POC, Clinic, Labs, MRI Machines,
Doctors’ Offices PDAs, Implants,
Healthcare and Life
Implants, Home, Surgical Equipment,
Science In Vivo/Home
Monitoring Systems Pumps, Monitors,
Drug Discovery, Telemedicine, etc.
Research
Diagnostics, Labs
Non-Vehicular Air, Rail, Marine
Consumer, Vehicles, Lights,
Transportation Commercial, Ships, Planes,
Vehicles Signage, Tolls, etc.
Construction, Off-
Highway
Tolls, Traffic
Transport Systems Management,
Navigation
Mining, Irrigation,
Resource
Agricultural,
Automation
Woodland
Petrochemicals,
Fluid/Processes Hydro, Carbons, Pumps, Valves, Vats,
Food, Beverages Conveyors,
Industrial Metals, Papers, Fabrication,
Rubber/Plastic, Assembly/Packaging,
Converting/ Discrete Metalworking, Vessels/Tanks, etc.
Electronics,
Assembly/Test
Pipelines,
Distribution
Conveyance
Fuel Stations,
Gaming, Bowling,
Specialty
Cinemas, Discos,
Special Events POS Terminals,
Hotels, Restaurants, Tags, Cash Registers,
Retail Hospitality
Bars, Cafes, Clubs Vending Machines,
Supermarkets, Signs, etc.
Shopping Centers,
Stores
Single Site,
Distribution Centers
Radar/Satellite,
Environment,
Surveillance
Military Security,
Unmanned, Fixed
Weapons, Vehicles,
Equipment Tanks, Fighter Jets,
Ships, Aircraft, Gear
Battlefields, Jeeps,
Human, Animal, Cars, Ambulance,
Security/Public
Tracking Postal, Food, Health, Homeland Security,
Safety
Baggage Environment,
Water, Treatment, Monitor, etc.
Building,
Environment,
Public Infrastructure
Equipment and
Personnel, Police,
Fire, Regulatory
Ambulance, Police,
Emergency Services Fire, Homeland
Security
Services, E-
Commerce, Data
Public Servers, Storage,
Centers, Mobile
IT and Networks PCs, Routers,
Carriers, ISPs
Switches, PBXs, etc.
IT/Data Center
Private Enterprise
Office, Privacy Nets
Table 18-01: IoT Application Areas and Devices

IoT Technologies and Protocols


The Internet of Things (IoT) encompasses a broad spectrum of emerging technologies and
competencies. However, a significant challenge in the IoT space is the underdeveloped nature
of many technologies and the services provided by vendors, which complicates their effective
deployment in organizations. IoT relies heavily on standardized networking protocols to ensure
successful communication between endpoints. Below are key communication technologies and
protocols based on the distance between the source and destination:
Short-Range Wireless Communication
● Bluetooth Low Energy (BLE): BLE, also known as Bluetooth Smart, is a wireless personal
area network commonly used in sectors such as healthcare, security, fitness, and
entertainment.
● Light-Fidelity (Li-Fi): Similar to Wi-Fi but with two main differences—communication
mode and speed—Li-Fi uses Visible Light Communication (VLC) to transfer data through
household light bulbs at speeds of up to 224 Gbps.
● Near-Field Communication (NFC): NFC is a short-range communication technology that
uses magnetic induction to facilitate communication between electronic devices. It is
commonly applied in mobile payments, social media, and product identification.
● QR Codes and Barcodes: Machine-readable codes containing product or item information.
QR codes are two-dimensional and scannable by smartphones, while barcodes can be one-
dimensional or two-dimensional.
● Radio-Frequency Identification (RFID): RFID tags store data and are read through
electromagnetic fields. This technology is used in various industrial, healthcare, logistics,
and animal tracking sectors.
● Thread: An IPv6-based protocol designed for home automation, allowing IoT devices to
communicate with one another on local wireless networks.
● Wi-Fi: It is a commonly used wireless networking technology for Local Area Networks
(LANs). The 802.11n standard offers speeds up to 600 Mbps and a range of about 50 meters.
● Wi-Fi Direct: Enables peer-to-peer communication between devices without a wireless
access point. Devices establish communication by selecting one as the access point.
● Z-Wave: A low-power, short-range communication protocol mainly for home automation,
allowing devices like thermostats and home security systems to communicate wirelessly.
● Zigbee: Based on the IEEE 203.15.4 standard, Zigbee is designed for low data-rate
communication over short ranges of 10-100 meters, typically used in devices with infrequent
data transmission needs.
● ANT: Adaptive Network Topology (ANT) is a wireless sensor network technology for short-
range communication in sports and fitness devices.
Medium-Range Wireless Communication
● HaLow: A variant of the Wi-Fi standard, HaLow is designed to provide extended coverage,
especially in rural areas. It operates at lower data rates, which helps reduce power
consumption and transmission costs.
● LTE-Advanced: An improved version of LTE, LTE-Advanced enhances mobile
communication by offering greater data capacity, extended coverage, better efficiency, and
improved performance.
● 6LoWPAN: IPv6 over Low-Power Wireless Personal Area Networks (6LoWPAN) is a
protocol that enables communication between small, low-power devices with limited
processing abilities, such as various IoT devices.
● QUIC: Quick UDP Internet Connections (QUIC) are multiplexed connections that allow
IoT devices to communicate over the User Datagram Protocol (UDP), providing security
features comparable to SSL/TLS.
Long-Range Wireless Communication
● LPWAN: Low Power Wide Area Network (LPWAN) is a wireless communication technology
that enables long-distance communication between two endpoints. Some common LPWAN
protocols and technologies include:
o LoRaWAN: A Long Range Wide Area Network (LoRaWAN) is used for applications such
as mobile communication, industrial machine-to-machine exchanges, and secure two-
way communications, particularly in IoT, smart cities, and healthcare.
o Sigfox: This technology is used for devices that require minimal data transfer and have
short battery lifespans.
o Neul: Operating in a small section of the TV white space spectrum, Neul provides high-
quality, high-power, wide-coverage, and cost-effective network solutions.
● VSAT: Very Small Aperture Terminal (VSAT) is a communication protocol utilizing small
dish antennas for both broadband and narrowband data transmission.
● Cellular: Cellular communication is used for long-distance data transfer. It offers high-
quality transmission but has the drawbacks of higher costs and significant power
consumption.
● MQTT: Message Queuing Telemetry Transport (MQTT) is a lightweight ISO standard
protocol for sending messages over long-range wireless networks. It is often utilized in
remote locations such as satellite links.
● NB-IoT: Narrowband IoT (NB-IoT) is an advanced version of LoRaWAN and Sigfox, using
enhanced physical layer technology and a dedicated spectrum for machine-to-machine
communication.
Wired Communication
● Ethernet: Ethernet is today's most widely used networking protocol, commonly employed
in Local Area Networks (LANs). It connects computers within a limited area, such as a
building, office, or campus, through wired connections.
● Multimedia over Coax Alliance (MoCA): MoCA is a networking protocol that delivers
high-definition video and related content to homes using existing coaxial cables.
● Power-Line Communication (PLC): PLC is a protocol that enables the transmission of
both power and data through electrical wiring. It is essential for various applications,
including home automation, industrial systems, and Broadband over Power Lines (BPL).
IoT Operating Systems
IoT devices consist of both hardware and software elements. The hardware includes
components like end devices and gateways, while the software typically involves operating
systems. As the production of hardware components such as sensor nodes and gateways has
increased, traditional IoT devices, which previously functioned without an operating system,
have started to adopt new OS implementations tailored specifically for IoT applications. These
operating systems enhance device connectivity, usability, and interoperability. Below are some
operating systems used in IoT devices:
 Windows 10 IoT: A family of operating systems developed by Microsoft, designed for
embedded systems.
 Amazon FreeRTOS: It is an open-source, free OS that enables easy deployment,
security, connectivity, and management of low-power, battery-operated IoT edge
devices.
 Fuchsia: A Google-developed open-source OS that supports various platforms,
including embedded systems, smartphones, and tablets.
 RIOT: A lightweight OS that uses minimal resources and is energy-efficient, suitable for
embedded systems, sensors, and actuator boards.
 Ubuntu Core: Also known as Snappy, used in devices like robots, drones, and edge
gateways.
 ARM Mbed OS: Typically used in low-power devices such as wearables.
 Zephyr: Designed for low-power, resource-constrained devices.
 Embedded Linux: Used across small, medium, and large embedded systems.
 NuttX RTOS: An open-source OS that supports 8-bit and 32-bit microcontrollers in
embedded systems.
 Integrity RTOS: Primarily used in aerospace, defense, automotive, and medical
industries.
 Apache Mynewt: Supports devices utilizing the Bluetooth Low Energy (BLE) protocol.
 Tizen: An open-source, Linux-based OS used in various devices, such as smartphones,
tablets, smart TVs, wearables, and IoT devices.
IoT Application Protocols
● Constrained Application Protocol (CoAP): This protocol is designed for transferring
messages between IoT networks and constrained nodes. It is primarily used in machine-to-
machine (M2M) applications, such as building automation and smart energy systems.
● Edge Computing: Edge computing shifts computational tasks closer to the network's edge,
allowing smart devices and gateways to perform services and tasks from the cloud. This
decentralization enhances content caching, delivery, storage, and overall management in
IoT networks.
● Lightweight Machine-to-Machine (LWM2M): LWM2M is a communication protocol at
the application layer that facilitates communication between IoT devices and assists in their
management.
● Physical Web: This technology enables quick and seamless interaction with nearby IoT
devices by broadcasting a list of URLs from devices via Bluetooth Low Energy (BLE) beacons.
● eXtensible Messaging and Presence Protocol (XMPP): XMPP is an open-source
protocol for real-time communication used to develop interoperable devices, applications,
and services within the IoT ecosystem.
● Mihini/M3DA: Mihini/M3DA is a software framework that facilitates communication
between M2M servers and embedded gateway applications. It allows IoT applications to
exchange data and control commands with M2M servers.
IoT Communication Models
IoT technology utilizes several communication models, each offering distinct features that allow
devices to interact with one another or a client. Below are four communication models and their
key characteristics:
● Device-to-Device Communication Model
This model connects devices through the internet, typically using protocols like ZigBee, Z-
Wave, or Bluetooth. It is commonly found in smart home products such as thermostats, lights,
door locks, CCTV cameras, and refrigerators, where devices exchange small data packets at low
data rates. This model is also used in wearable devices. For instance, an ECG/EKG device
attached to a patient can sync with their smartphone to send emergency notifications.

Figure 18-03: IoT Device-to-Device Communication Model

● Device-to-Cloud Communication Model


In this model, devices connect to the cloud for data exchange or command transmission instead
of directly communicating with the client. It typically relies on protocols like Wi-Fi or Ethernet
and sometimes cellular networks. For example, a Wi-Fi-enabled CCTV camera illustrates this
model. The camera cannot communicate directly with the user’s smartphone; instead, it
uploads data to the cloud. Upon providing the correct credentials, the user gains access to the
cloud, allowing them to view the camera’s feed remotely.
Figure 18-04: IoT Device-to-Cloud Communication Model

● Device-to-Gateway Communication Model


In this model, the IoT device communicates with a gateway, relaying data to the cloud. The
gateway is an intermediary, often providing security features and handling data or protocol
conversion. Common protocols used in this model include ZigBee and Z-Wave.
For example, if a smartphone serves as the gateway, it may run an app that connects to the IoT
device and the cloud. A smart TV, for instance, could connect to a cloud service through a
mobile app acting as the gateway.

Figure 18-05: IoT Device-to-Gateway Communication Model

● Back-End Data-Sharing Communication Model


This model builds on device-to-cloud communication by allowing authorized third parties to
access data uploaded by IoT devices to the cloud. The data is stored in the cloud and later
retrieved or analyzed by these third parties. For instance, an energy consumption analyzer
might assess a company's energy usage monthly or yearly. The insights gained can help the
company implement strategies to reduce energy costs through energy-saving or harvesting
techniques.
Figure 18-06: IoT Back-End Data-Sharing Communication Model

Challenges of IoT
The rapid expansion of IoT technology has made it an integral part of daily life, with numerous
applications and features. However, the lack of robust security policies makes IoT devices
vulnerable to cyberattacks. New updates to these devices also often introduce security gaps that
can be exploited. To address these concerns, manufacturers must prioritize security from the
planning and design phases through deployment, implementation, management, and
maintenance. Below are some key challenges that make IoT devices susceptible to threats:
● Inadequate Security and Privacy: Many IoT devices, such as home appliances, industrial
tools, healthcare devices, and vehicles, are connected to the internet and store sensitive
data. However, they often lack fundamental security measures, leaving them open to
exploitation by hackers.
● Weak Web Interfaces: Many IoT devices have built-in web server technology, which can
make them vulnerable to attacks if they are not properly secured.
● Regulatory and Legal Issues: The interconnected nature of IoT devices raises security
concerns, but there is often a lack of laws to address these challenges effectively.
● Default, Weak, and Hardcoded Credentials: A common vulnerability in IoT devices is
using default and easily guessable credentials. These weak authentication systems are prime
targets for hackers seeking unauthorized access.
● Unencrypted Data Transmission and Open Ports: IoT devices often transmit data in
clear text and have open ports, exposing them to interception or exploitation, especially
since they usually lack encryption during data transfer.
● Programming Vulnerabilities (Buffer Overflow): Many IoT devices include embedded
web services vulnerable to common attacks such as buffer overflows and SQL injection.
These can be exploited when the device functionality is updated.
● Storage Limitations: IoT devices generally have limited storage capacity, yet they gather
large amounts of data. This discrepancy leads to challenges in managing, storing, and
securing the data effectively.
● Challenges in Updating Firmware and OS: Firmware updates are crucial for securing
devices but may affect performance. Manufacturers may be reluctant to release updates due
to the potential impact on functionality.
● Interoperability Issues: IoT devices often struggle with compatibility, making it difficult
to test APIs, secure devices with third-party software, or manage and monitor devices across
a unified platform. This lack of interoperability hinders the growth and effectiveness of the
IoT ecosystem.
● Physical Theft and Tampering: IoT devices are vulnerable to physical attacks, such as
tampering or counterfeiting, which allow attackers to inject malicious code or alter the
hardware.
● Limited Vendor Support for Vulnerability Patches: Device manufacturers may be
hesitant or unwilling to grant third-party access to fix security vulnerabilities, leaving
devices at risk.
● Economic and Developmental Challenges: The rapid spread of IoT devices adds
complexity for policymakers, who must create new regulations and frameworks to address
the emerging issues surrounding these technologies.
● Management of Unstructured Data: The increasing number of connected devices
generates vast amounts of unstructured data. As this data's volume, velocity, and variety
grow, organizations must figure out how to identify and handle valuable, actionable
information.
● Scalability: As the IoT network expands, managing and scaling the infrastructure becomes
more complex, requiring improved data management and analysis capabilities.
● Energy Consumption: Many IoT devices are battery-powered, making optimizing energy
use challenging to extend their operational life.
● Compliance with Regulations: IoT deployments must comply with various regulations
related to data security, privacy, and protection, which differ across regions and industries.
● Integration with Legacy Systems: Integrating IoT technology with existing infrastructure
and systems can be difficult, requiring smooth compatibility and seamless integration to
avoid disruptions.
Threat vs. Opportunities
If the IoT is not properly configured or understood, it can present significant risks to personal
data, privacy, and safety. However, when appropriately managed and secured, IoT can enhance
communication, service delivery, and overall quality of life. The threats posed by IoT can be
grouped into three main areas: security, privacy, and safety. These areas are interconnected as
they relate to the same device and its network connectivity. The significance of these concerns
is growing as IoT devices are becoming more integrated into daily life than smartphones, and
they can access highly sensitive personal data, including health records, financial information,
and social security numbers. While concerns with smartphones and tablets may be limited, the
potential risks with IoT devices are far greater. Therefore, safeguarding these devices' security,
privacy, and safety is critical. Addressing these three threat categories through effective
strategies will lead to more secure communication, reduced cyberattacks, better user
experiences, and increased cost efficiency.
IoT Security Problems
IoT systems can present significant risks to organizations due to potential vulnerabilities. Many
IoT devices are prone to security issues, including lack of adequate authentication methods,
reliance on default credentials, absence of lock-out features, weak encryption protocols, poor
key management, and inadequate physical security. Figure 18-07 shows some of the security
concerns associated with each layer of the IoT architecture:
Figure 18-07: Security Problems in IoT Architecture

OWASP Top 10 IoT Threats


The Open Web Application Security Project (OWASP) identifies the following as the top 10
threats to IoT systems:
1. Weak, Guessable, or Hardcoded Passwords
Brute-force attacks can exploit easily guessed or hardcoded passwords, including publicly
available or unchangeable credentials. Backdoors in device firmware or client software further
enable unauthorized access.
2. Insecure Network Services
Unsecured network services are vulnerable to attacks such as buffer overflows, leading to denial-
of-service scenarios that render devices inaccessible. Attackers can exploit open ports using
automated tools like port scanners and fuzzers, compromising data confidentiality,
authenticity, and availability while enabling unauthorized remote access.
3. Insecure Ecosystem Interfaces
Vulnerabilities in interfaces such as web portals, backend APIs, mobile apps, or cloud systems
can jeopardize device security. Common flaws include weak or missing authentication, poor
encryption, and insufficient input/output validation.
4. Lack of Secure Update Mechanisms
Failure to implement secure update processes—such as firmware validation, anti-rollback
measures, or secure delivery—leaves devices open to attacks, including tampering and
exploitation during updates.
5. Use of Insecure or Outdated Components
Utilizing outdated software libraries, insecure OS customizations, or third-party components
from compromised supply chains can expose devices to significant risks, leading to breaches
and unauthorized access.
6. Insufficient Privacy Protection
Devices with inadequate privacy safeguards may expose sensitive personal information stored
within the system or its ecosystem to attackers.
7. Insecure Data Transfer and Storage
Lack of proper encryption and access controls for data in transit or at rest can result in sensitive
information being intercepted or leaked to malicious actors.
8. Lack of Device Management
The absence of robust device management features—such as asset tracking, secure updates,
monitoring, and secure decommissioning—creates vulnerabilities that attackers can exploit.
9. Insecure Default Settings
Default device configurations often lack adequate security, restricting operators from
implementing safer settings and exposing devices to attacks.
10. Lack of Physical Hardening
Insufficient physical security measures can allow attackers to extract sensitive information,
facilitating remote attacks or enabling direct control over the device.
OWASP IoT Attack Surface Areas
The following are the key IoT attack surface areas identified by OWASP:

Attack Surface Area Vulnerabilities

● Interoperability standards
● Data governance
● System-wide failure
1. Ecosystem ● Individual stakeholder risks
(General) ● Implicit trust between components
● Enrollment security
● Decommissioning system
● Lost access procedures
● Sensitive data
o Cleartext usernames
2. Device Memory o Cleartext passwords
o Third-party credentials
o Encryption keys
● Firmware extraction
● User CLI
● Admin CLI
● Privilege escalation
3. Device Physical ● Reset to an insecure state
Interfaces ● Removal of storage media
● Tamper resistance
● Debug port
o UART (Serial)
o JTAG/SWD
● Device ID/serial number exposure
● Standard set of web application vulnerabilities:
o OWASP Web Top 10
o OWASP ASVS
o OWASP Testing Guide
4. Device Web ● Credential management vulnerabilities:
Interface o Username enumeration
o Weak passwords
o Account lockout
o Known default credentials
o Insecure password recovery mechanism
● Sensitive data exposure
o Backdoor accounts
o Hardcoded credentials
o Encryption keys
o Encryption (symmetric, asymmetric)
o Sensitive information
5. Device Firmware o Sensitive URL disclosure
● Firmware version display and/or date of last update
● Vulnerable services (web, SSH, TFTP, etc.)
o Verify for old software versions and possible attacks
(Heartbleed, Shellshock, old PHP versions, etc.)
● Security-related function API exposure
● Firmware downgrade possibility
● Information disclosure
● User CLI
● Administrative CLI
● Injection
● Denial-of-Service
● Unencrypted services
● Poorly implemented encryption
● Test/development services
● Buffer overflow
● UPnP
6. Device Network
● Vulnerable UDP services
Services
● Device firmware OTA update block
● Firmware loaded over an insecure channel (no TLS)
● Replay attack
● Lack of payload verification
● Lack of message integrity check
● Credential management vulnerabilities:
o Username enumeration
o Weak passwords
o Account lockout
o Known default credentials
o Insecure password recovery mechanism
● Standard set of web application vulnerabilities:
o OWASP Web Top 10
o OWASP ASVS
o OWASP Testing Guide

● Credential management vulnerabilities:


o Username enumeration
7. Administrative o Weak passwords
Interface o Account lockout
o Known default credentials
o Insecure password recovery mechanism
● Security/encryption options
● Logging options
● Two-factor authentication
● Check for insecure direct object references
● Inability to wipe a device
● Unencrypted data
● Data encrypted with discovered keys
8. Local Data Storage
● Lack of data integrity checks
● Use of static same encryption/decryption key
● Standard set of web application vulnerabilities:
o OWASP Web Top 10
o OWASP ASVS
o OWASP Testing Guide

● Credential management vulnerabilities:


9. Cloud Web Interface o Username enumeration
o Weak passwords
o Account lockout
o Known default credentials
o Insecure password recovery mechanism
● Transport encryption
● Two-factor authentication
● Unencrypted PII sent
10. Third-party ● Encrypted PII sent
Backend APIs ● Device information leaked
● Location leaked
● Update sent without encryption
● Updates not signed
● Update location writable
11. Update Mechanism
● Update Verification
● Update authentication
● Malicious update
● Missing update mechanism
● No manual update mechanism
● Implicitly trusted by device or cloud
● Username enumeration
● Account lockout
● Known default credentials
12. Mobile Application ● Weak passwords
● Insecure data storage
● Transport encryption
● Insecure password recovery mechanism
● Two-factor authentication
● Inherent trust in the cloud or mobile application
● Weak authentication
13. Vendor Backend
● Weak access controls
APIs
● Injection attacks
● Hidden services
● Health checks
● Heartbeats
14. Ecosystem
● Ecosystem commands
Communication
● Deprovisioning
● Pushing updates
● LAN
● LAN to internet
● Short range
15. Network Traffic
● Non-standard
● Wireless (Wi-Fi, Z-wave, XBee, Zigbee, Bluetooth, LoRa)
● Protocol fuzzing
● Authentication/authorization-related values (session key,
token, cookie, etc.) disclosure
● Reuse of session key, token, etc.
● Device-to-device authentication
16. Authentication/Aut
● Device-to-mobile-application authentication
horization
● Device-to-cloud-system authentication
● Mobile-application-to-cloud-system authentication Web-
application-to-cloud-system authentication
● Lack of dynamic authentication
● User data disclosure
17. Privacy ● User/device location disclosure
● Differential privacy
● Sensing environment manipulation
18. Hardware (Sensors) ● Tampering (physical)
● Damage (physical)
Table 18-02: OWASP IoT Attack Surface Areas
IoT Vulnerabilities
Table 18-03 covers the vulnerabilities identified in IoT systems by OWASP:

Vulnerability Attack Surface Description

● Administrative Interface ● Ability to collect a set of valid


1. Username ● Device Web Interface usernames by interacting
Enumeration ● Cloud Interface with the authentication
● Mobile Application mechanism
● Ability to set account
● Administrative Interface
passwords to “1234” or
● Device Web Interface
2. Weak Passwords “123456”, for example
● Cloud Interface
● Usage of pre-programmed
● Mobile Application
default passwords
● Administrative Interface ● Ability to continue sending
● Device Web Interface authentication attempts after
3. Account Lockout
● Cloud Interface 3–5 failed login attempts
● Mobile Application
● Network services are not
4. Unencrypted properly encrypted to
● Device Network Services
Services prevent eavesdropping or
tampering by attackers
● Lack of two-factor
● Administrative Interface
5. Two-factor authentication mechanisms
● Cloud Web Interface
Authentication such as a security token or
● Mobile Application
fingerprint scanner
● Encryption is implemented;
6. Poorly however, it is improperly
Implemented ● Device Network Services configured or is not being
Encryption properly updated, e.g., using
SSL v2
● Updates are transmitted over
7. Update Sent the network without using
● Update Mechanism
Without Encryption TLS or encrypting the update
file itself
● Storage location for update
files is world-writable, which
8. Update Location
● Update Mechanism can allow firmware to be
Writable
modified and distributed to
all users
● Service can be attacked in a
way that denies service to
9. Denial-of-Service ● Device Network Services
that service or the entire
device
● Ability to physically remove
10. Removal of Storage ● Device Physical
the storage media from the
Media Interfaces
device
● No ability to manually force
11. No Manual Update
● Update Mechanism an update check for the
Mechanism
device
12. Missing Update ● No ability to update the
● Update Mechanism
Mechanism device
● Current firmware version is
13. Firmware Version
not displayed, and/or the
Display and/or Last ● Device Firmware
date of the last update is not
Update Date
displayed
● JTAG/SWD Interface ● Firmware contains a lot of
● In-Situ Dumping useful information, like
● Intercepting an Over-the- source code and binaries of
Air (OTA) Update running services, preset
14. Firmware and ● Downloading from the passwords, and SSH keys
Storage Extraction Manufacturer’s Web Page
● eMMC Tapping
● Unsoldering the SPI
Flash/eMMC Chip and
Reading It in an Adapter
● With the help of a JTAG
adapter and GNU debugger,
we can modify the execution
of firmware in the device and
15. Manipulating the ● JTAG/SWD Interface
bypass almost all software-
Code Execution Flow Side-Channel Attacks like
based security controls
of the Device Glitching
● Side-channel attacks can also
modify the execution flow or
can be used to leak
interesting
● By connecting to a serial
interface, we can obtain full
console access to a device
● Usually, security measures
16. Obtaining Console ● Serial Interfaces
include custom bootloaders
Access (SPI/UART)
that prevent the attacker
from entering single-user
mode, but that can also be
bypassed
● Out-of-date versions of
17. Insecure Third-
● Software BusyBox, OpenSSL, SSH, web
Party Components
servers, etc.
Table 18-03: IoT Vulnerabilities

IoT Threats
IoT devices often lack robust security mechanisms, making them vulnerable to emerging
threats. These devices can quickly become compromised by malware or malicious code,
enabling attackers to exploit them for harmful purposes such as disrupting networks,
intercepting communications, and executing large-scale attacks like Distributed Denial-of-
Service (DDoS). Below are some common types of IoT attacks:
● DDoS Attack: Devices are hijacked and turned into a network of botnets to overload a
target system or server, rendering it incapable of delivering services.
● HVAC System Attack: Attackers exploit weaknesses in HVAC systems to steal sensitive
information, such as credentials, and launch additional attacks on the network.
● Rolling Code Attack: Criminals intercept and replicate the signal to unlock vehicles,
enabling unauthorized access and theft.
● BlueBorne Attack: Exploiting vulnerabilities in Bluetooth protocols, attackers connect to
nearby devices to compromise their functionality and data.
● Jamming Attack: Malicious traffic is introduced to disrupt communication between sender
and receiver, making the exchange of information impossible.
● Backdoor Exploitation: Vulnerabilities in IoT devices are leveraged to create backdoors,
granting attackers unauthorized access to organizational networks.
● Telnet Exploitation: Open telnet ports are targeted to extract sensitive information shared
among devices, including their software and hardware configurations.
● Sybil Attack: Fake identities are generated to simulate traffic congestion, hindering
effective communication between nodes in a network.
● Exploit Kits: Malicious scripts target unpatched vulnerabilities in IoT devices,
compromising their security.
● Man-in-the-Middle Attack: An attacker impersonates a legitimate party, intercepting and
hijacking communication between two endpoints.
● Replay Attack: Authentic messages are intercepted and repeatedly sent to target devices to
cause denial-of-service or crashes.
● Forged Malicious Device: Genuine IoT devices are replaced with malicious ones by
attackers with physical access to the network.
● Side-Channel Attack: Information about encryption keys is extracted from IoT devices by
monitoring emissions (e.g., signals or power usage).
● Ransomware Attack: Malware encrypts or locks a user’s device, preventing access until a
ransom is paid.
● Client Impersonation: Attackers pose as legitimate devices or servers to access sensitive
information or perform unauthorized actions.
● SQL Injection Attack: Weaknesses in mobile or web applications managing IoT devices
are exploited to gain access and execute further attacks.
● SDR-Based Attack: Using software-defined radio systems, attackers analyze
communication signals and send spam messages to connected devices.
● Fault Injection Attack: Malicious faults are introduced into IoT devices, allowing attackers
to exploit these vulnerabilities and compromise security.
● Network Pivoting: Attackers use compromised devices to infiltrate a secure server and
then leverage this access to breach other networked systems.
● DNS Rebinding Attack: Malicious JavaScript code is used to gain access to a victim’s router,
facilitating unauthorized activities.
● Firmware Update Attack (FOTA): Attackers tamper with the firmware update process to
inject harmful code into devices.
Hacking IoT Devices: General Scenario
The Internet of Things (IoT) encompasses a range of technologies, including embedded sensors,
microprocessors, and power management systems. Security requirements can vary significantly
depending on the device and its application. As the volume of sensitive data transmitted across
networks increases, so does the potential for risks such as data breaches, manipulation,
tampering, and attacks targeting routers and servers. A weak security framework can result in
the following issues:
● Interception of Communication: An unauthorized individual may intercept data
exchanged between two endpoints, gaining access to sensitive information. This stolen data
can then be misused for personal gain.
● Deployment of Fake Servers: Malicious actors could deploy counterfeit servers to send
unauthorized commands, triggering unintended actions. For instance, valuable resources
like water, electricity, or oil might be rerouted to unauthorized destinations.
● Injection of Malicious Code: Fake devices could introduce harmful scripts into a system,
forcing it to operate in unsafe or unintended ways, potentially leading to hazardous
outcomes.

Figure 18-08: General IoT Device Hacking Scenarios

DDoS Attack
A Distributed Denial-of-Service (DDoS) attack is a malicious activity in which multiple
compromised systems overwhelm a single online system or service, temporarily rendering it
slow, unresponsive, or inaccessible to legitimate users. The process begins with the attacker
exploiting device vulnerabilities and installing malicious software on their operating systems.
These compromised devices, collectively called a botnet, act as agents for the attack.
Once the attacker selects a target, they command the botnet or zombie agents to flood the target
server with overwhelming requests. These requests originate from multiple IoT devices across
various regions, overloading the target system with more traffic than it can manage.
Consequently, the system may experience degraded performance, go offline, or completely shut
down.
The typical steps an attacker follows to execute a DDoS attack on IoT devices are:
1. Gaining remote access to vulnerable IoT devices.
2. Injecting malware into the devices to convert them into botnets.
3. Using a command-and-control center to direct botnets to send massive requests to the
target server.
4. Overwhelming the target server, causing it to go offline and not process legitimate requests.

Figure 18-09: DDoS Attack on IoT Devices

Exploit HVAC
Many organizations deploy internet-connected Heating, Ventilation, and Air Conditioning
(HVAC) systems without adequate security measures, creating potential entry points for
attackers to infiltrate corporate systems. These systems often have vulnerabilities that
cybercriminals exploit to steal login credentials, gain control of the HVAC system, and launch
additional attacks on the organization's network.
HVAC systems are commonly integrated into networks across industries, government
institutions, hospitals, and other sectors. They often allow remote access for vendors and third
parties to perform administrative tasks, such as monitoring energy usage and adjusting
temperatures remotely. Unfortunately, many HVAC providers issue standardized login
credentials to multiple organizations, making it easier for attackers to gain unauthorized access.
Once inside, they can exploit these systems to infiltrate corporate networks and extract sensitive
data.
The typical steps an attacker follows to compromise HVAC systems include:
1. Using platforms like Shodan (https://www.shodan.io) to locate vulnerable Industrial
Control Systems (ICSs).
2. Searching for default user credentials through resources like https://www.defpass.com.
3. Attempting to log in to the ICS using these default credentials.
4. Gaining remote access to the HVAC system via the ICS.
5. Exploiting the HVAC system to manipulate settings, such as temperature controls, or
conducting further attacks within the local network.

Figure 18-10: Exploiting HVAC System

Rolling Code Attack


Many modern vehicles utilize advanced locking mechanisms that rely on RF signals transmitted
as unique codes from a key fob to lock or unlock the vehicle. These codes, known as rolling or
hopping codes, change with each use and are valid only once. This ensures that if the same code
is received again, the system rejects it, offering protection against replay attacks.
Despite these precautions, attackers can exploit vulnerabilities in the system to capture and
misuse rolling codes. Using a jamming device, they employ a method that involves interrupting
the signal transmission from the key fob to the vehicle’s receiver. This device simultaneously
blocks the signal and captures the transmitted code. The attacker can then use the stolen code
to unlock the vehicle or garage door later.
The typical steps in a rolling code attack include:
1. The victim presses the key fob to unlock the car.
2. The attacker uses a jamming device to block the vehicle from receiving the rolling code and
simultaneously intercepts the first code.
3. The vehicle remains locked, prompting the victim to press the key fob again, transmitting a
second code.
4. The attacker intercepts the second code while replaying the first code to unlock the vehicle.
5. The attacker later uses the intercepted second code to gain unauthorized access to the car
or garage.
Attackers commonly utilize tools such as HackRF One and RFCrack to execute such exploits.

Figure 18-11: Illustration of Rolling Code Attack

BlueBorne Attack
A BlueBorne attack exploits Bluetooth connections to gain unauthorized access and take full
control of a target device. By leveraging vulnerabilities in the Bluetooth protocol, attackers can
connect to nearby devices and compromise them. BlueBorne is a set of techniques based on
identified flaws in the Bluetooth protocol. It can target a wide range of IoT devices running
operating systems like Android, Linux, Windows, and older versions of iOS.
The Bluetooth process in these systems generally has elevated privileges, making it an attractive
target. Once attackers gain access to a single device, they can infiltrate corporate networks, steal
sensitive data, and distribute malware to nearby devices. BlueBorne attacks are effective across
various software versions and require no user interaction, prior configuration, or pairing—
Bluetooth needs to be active.
This attack allows an intruder to identify Bluetooth-enabled devices, even those not in discovery
mode. Once a device is detected, the attacker can extract its MAC address and operating system
details to exploit system-specific vulnerabilities. Depending on the flaws in the Bluetooth
protocol, attackers can execute malicious code remotely or launch man-in-the-middle attacks
to compromise the target.
Smartphones, smart TVs, wearable devices, car audio systems, printers, and other devices are
susceptible to BlueBorne attacks.
Steps involved in a BlueBorne attack:
1. Discovery: The attacker scans for Bluetooth-enabled devices within range, even those not
in discoverable mode.
2. MAC Address Retrieval: Once a device is located, the attacker extracts its MAC address.
3. Operating System Identification: The attacker repeatedly probes the target device to
determine its operating system.
4. Exploitation: Based on identified vulnerabilities in the Bluetooth protocol, the attacker
exploits the target OS to gain access.
5. Execution: The attacker performs remote code execution or a man-in-the-middle attack,
gaining complete control of the device.

Figure 18-12: Illustration of BlueBorne Attack

Jamming Attack
Jamming is a form of attack that disrupts communication between wireless IoT devices,
effectively compromising their functionality. This attack involves flooding the communication
channel with excessive malicious traffic, leading to a Denial-of-Service (DoS) condition for
legitimate users. As a result, authorized devices cannot exchange data, and communication is
obstructed.
All wireless devices and networks are vulnerable to jamming attacks. Attackers employ
specialized hardware to emit random radio signals at the same frequency as the target device's
communication channel. Wireless devices perceive these jamming signals as noise, prompting
them to suspend their transmissions until the interference ceases. This continuous noise
overloads the network, preventing devices from sending or receiving data and causing a
complete communication breakdown.
Figure 18-13: Illustration of Jamming Attack

Hacking Smart Grid/ Industrial Devices: Remote Access Using Backdoor


Attackers use social engineering tactics to collect essential information about their target
organization. Once they obtain details such as employees' email addresses, they send phishing
emails containing malicious attachments, like a word document. When an employee opens the
email and clicks the attachment, a backdoor is silently installed on the system. This backdoor
gives the attacker unauthorized access to the organization's internal network.
For instance, in an attack targeting a power grid, the attacker can access the organization's
private network and then infiltrate the Supervisory Control and Data Acquisition (SCADA)
system that manages the grid. After compromising the SCADA network, the attacker replaces
the legitimate firmware with malicious versions, enabling them to manipulate the system.
Ultimately, the attacker can disrupt the power supply by sending harmful commands to the
substation control systems through the SCADA network.

Figure 18-14: Hacking a Smart Grid to Gain Remote Access


SDR-Based Attacks on IoT
Software-Defined Radio (SDR) is a technique for generating radio communications and
performing signal processing through software (or firmware) instead of traditional hardware-
based methods. Using SDR systems, an attacker can monitor communication signals within IoT
networks and transmit unwanted messages or data to connected devices. SDR technology also
enables manipulation of the signal transmission and reception between devices, depending on
how the software is implemented. This type of attack can be carried out in full-duplex (two-way
communication) and half-duplex (one-way communication) modes.
Replay Attack
One common SDR-based attack in IoT environments is the replay attack. In this attack, the
attacker intercepts a sequence of commands sent between devices and later replays them to
gain unauthorized access or control. The steps involved in executing a replay attack are as
follows:
 The attacker targets the specific frequency used for communication between devices.
 After identifying the frequency, the attacker intercepts the original command data sent by
the devices.
 Once the data is captured, the attacker uses tools like the Universal Radio Hacker (URH) to
extract and organize the captured commands.
 Finally, the attacker retransmits the extracted commands back into the network on the same
frequency, effectively replaying the captured signals or commands to the devices involved.
Cryptanalysis Attack
A cryptanalysis attack is a significant threat to IoT devices. It involves a process similar to a
replay attack, with the added step of reverse-engineering the protocol to recover the original
signal. To carry out this attack, the attacker needs a strong understanding of cryptography,
communication theory, and modulation techniques to eliminate noise from the signal.
Although more challenging to execute than a replay attack, skilled attackers may use various
tools and methods to break security.
Reconnaissance Attack
This attack complements the cryptanalysis attack by allowing the attacker to gather information
from the device's specifications. IoT devices that rely on RF signals must be certified by their
respective country’s authority, which usually results in the release of an analysis report for the
device. To bypass this, designers sometimes obscure identification marks on the chipset. In
response, attackers may use tools like multimeters to examine the chipset, identify key
components like ground pins, and discover the product ID. They can then compare this
information to the official analysis report.
Identifying and Accessing Local IoT Devices
An attacker can gain control over local IoT devices when a user from the network visits a
malicious website, typically disguised as an advertisement or other enticing content. Upon
visiting the harmful site, a malicious JavaScript code embedded in the page is activated.
Attackers generally use two methods to seize control of local IoT devices.
Discovering or Identifying the Local IoT Devices
The first step involves discovering or identifying the IoT devices in the local network. The
process for this is as follows:
1. The attacker obtains the local IP address through the malicious code.
2. The attacker sends requests to all devices on the network.
3. Active devices reply with a reset packet, while requests to inactive devices time out.
4. The attacker identifies all available devices by analyzing the responses.

Figure 18-15: Discovering the Local IoT Devices

Accessing the Local IoT Devices Using DNS Rebinding


DNS rebinding is a method by which an attacker gains access to a victim’s router through
malicious JavaScript injected into a web page. This enables the attacker to target any device that
uses default login credentials. After identifying the devices connected to the network, the
attacker exploits the system to control the local devices completely. Once the attacker has
information about the IoT devices on the network, they proceed with the following steps:
1. Use DNS rebinding tools, like Singularity of Origin, to check if the malicious code is
performing DNS rebinding on the discovered devices.
2. After successfully carrying out DNS rebinding, the attacker can take control of the local
IoT devices.
3. The attacker can further gather sensitive data, such as the UIDs and BSSIDs of local
access points, which help determine the targeted devices' geographic location.

Figure 18-16: DNS Rebinding Attack on Local IoT Devices

Once the attack is successfully executed, the attacker can bypass security measures and gain
access to applications running on the local IoT devices. Additionally, the attacker could initiate
random audio or video files across the devices' browsers.
Fault Injection Attacks
Fault injection or perturbation attacks occur when an attacker introduces a faulty or malicious
program into a system to breach its security. These attacks can be carried out using various
methods and classified as invasive or non-invasive. In non-invasive attacks, the attacker must
be close to the chip to manipulate the default program or data and extract sensitive information.
Invasive attacks, however, require the attacker to have physical access to the chip surface. Below
are several types of fault injection attacks:
● Optical, Electromagnetic Fault Injection (EMFI), and Body Bias Injection (BBI):
These attacks aim to introduce faults into devices by using lasers and electromagnetic
pulses, targeting analog components like Random Number Generators (RNGs) and applying
high-voltage pulses to disrupt the system and compromise its security.
● Power/Clock/Reset Glitching: These attacks involve injecting faults or glitches into the
power supply, which can trigger remote execution or skip important instructions. Attackers
may also target the clock network, which ensures synchronized signals across the chip.
● Frequency/Voltage Tampering: In these attacks, attackers manipulate the operating
conditions of a chip by adjusting the power supply levels and altering the clock frequency.
The goal is to induce faulty behavior in the chip and weaken the device’s security.
● Temperature Attacks: Attackers alter the chip's operating temperature, which changes the
chip’s environment and allows it to be executed under non-standard conditions.
Once faults are introduced using these methods, attackers can exploit the device’s faulty
behavior to steal sensitive data or disrupt its normal operation.
Other IoT Attacks
Sybil Attack
Vehicular communication systems ensure safe transportation by sharing critical safety
information and traffic updates. However, even Vehicular Ad-hoc Networks (VANETs) are
vulnerable to attacks. In a Sybil attack, an attacker uses multiple fake identities to create the
illusion of traffic congestion, disrupting communication between adjacent nodes and networks.
This type of attack, which can severely affect network performance, is considered one of the
most dangerous threats to VANETs. It undermines the network’s applications by falsely
simulating traffic overload. In a Sybil attack, a vehicle is made to appear at multiple locations
simultaneously. For instance, a node, referred to as Sybil node “X,” may either create a new
identity or steal an existing legitimate one. Normally, nodes “A” and “B” should only
communicate with each other. However, in this attack, node “X” impersonates other nodes and
interferes with communication. Using several fake identities, node “X” disrupts the network,
causing confusion and significant security risks.
Exploit Kits
Exploit kits are malicious tools attackers employ to exploit inadequately patched weaknesses in
IoT devices. These kits are designed to automatically adapt to new vulnerabilities by adding
updated exploitation methods and additional features. Once a vulnerability is identified, the
exploit kit delivers the appropriate exploit to install malware, which can then execute and
damage the device. These kits are particularly dangerous because they often remain undetected
in IoT environments, compromising devices and infrastructure and causing them to
malfunction.
Man-in-the-Middle Attack
In a man-in-the-middle attack, the attacker positions themselves between the sender and the
receiver, impersonates a legitimate sender, and intercepts the communication. IoT devices are
typically connected to networks and act as gateways to sensitive data, making them vulnerable
to this attack. Malicious actors can pose as legitimate senders and send harmful requests to take
control of the device. Devices such as IP cameras, routers, modems, and internet gateways often
have cryptographic weaknesses that make them susceptible to man-in-the-middle attacks.
Replay Attack
A replay attack occurs when attackers capture valid messages from a communication and
repeatedly send the intercepted messages to the target device. This can lead to a Denial-of-
Service (DoS) attack, message manipulation, or even cause the target device to crash. For
instance, in a replay attack targeting an IoT-controlled front door, the attacker records the
infrared signal used to unlock the door, reproduces it, and sends it again to unlock the door.
Forged Malicious Device Attack
When attackers gain physical access to a network, they can substitute genuine IoT devices with
counterfeit ones. These forged devices are hard to detect since they closely resemble legitimate
devices. However, they often contain backdoors that allow attackers to execute malicious
actions within the network.
Side-Channel Attack
In a side-channel attack, attackers gather information about encryption keys by monitoring
signals emitted by IoT devices, such as power usage or electromagnetic emissions. Every device
produces these signals during its internal operations. By closely analyzing the variations in
power consumption, attackers can access and replicate encryption keys without directly
tampering with the device. This method is advantageous because it is quick and requires
minimal effort to extract encryption keys. The data leaked through these side channels can also
facilitate other attack techniques, like power analysis or time-based attacks.
Ransomware Attack
Ransomware is a form of malware that locks the screen or encrypts files to prevent a user from
accessing their device. The user is unable to regain access until a ransom is paid. This type of
attack can occur in various ways, such as through unintentional downloads of malicious files,
malware, or software and sometimes via harmful advertisements (malvertisements).
The typical stages of a ransomware attack are as follows:
● Phase 1: The victim receives an email that seems to come from a trusted source containing
an attachment with a malicious file.
● Phase 2:
o The user opens the email and clicks the malicious file, downloading malware and
triggering legitimate processes like PowerShell, Vssadmin encryption, or cmd.exe. This
connects the device to the attacker's Command-and-Control (C2) server.
o The victim’s files are then encrypted.
● Phase 3: A ransom notification appears on the victim’s device, demanding payment in
money or Bitcoin to unlock the encrypted files.
IoT Attacks in Different Sectors
IoT technology is advancing across various sectors such as industry, healthcare, agriculture,
smart cities, security, and transportation. However, IoT technology's decentralized nature often
leads organizations to pay less attention to device security. Instead of compartmentalizing IoT
systems, suppliers tend to focus on identifying and exploiting vulnerabilities. Attackers can
target these weaknesses in IoT devices to initiate attacks, including DoS, jamming, MITM, and
Sybil attacks, allowing them to collect sensitive data and compromising privacy and
confidentiality.
Table 18-04 covers the different IoT sectors and the corresponding attacks associated with them:

Service Sectors Types of Attack Possible Consequences

Access Control: Gaining access to Loss of confidentiality and


the device availability
MITM Attack: Listening to the Loss of privacy and data
communication between two confidentiality
endpoints
DoS Attack: Flooding data streams Loss of data availability
with communication to deplete
system resources
Eavesdropping: Collecting Loss of data confidentiality
exchanged messages
Buildings
Control Hijacking Attack: Loss of data availability
Changing normal flow control of the
IoT device firmware by injecting
malicious code.
Reverse Engineering: Analyzing Loss of privacy and data
the device firmware to obtain confidentiality
sensitive data
Rube Goldberg Attack: Chained Loss of privacy and data
attack designed to exploit industrial availability
IoT device weaknesses
Access Control: Gaining physical or Loss of confidentiality and
remote access to the device availability
Reconnaissance: Engages with the Loss of privacy and data
target system to obtain information confidentiality
DoS Attack: Making service Loss of data availability
Energy/Industrial
unavailable for legitimate users by
flooding the system with
communication requests
Eavesdropping: Collecting the Loss of data confidentiality
transmitted information
Rube Goldberg Attack: Chained Loss of privacy and data
attack designed to exploit availability
weaknesses of industrial IoT device
Spear Phishing Attack: Targets Loss of privacy and data
specific individuals or groups within confidentiality
an organization
Bluebugging: Exploiting Loss of privacy and data
vulnerabilities in old devices’ confidentiality
firmware and spying on phone calls
DoS Attack: Making service Loss of data availability
unavailable for legitimate users by
flooding the system with
communication requests
Access Control: Gaining access to Loss of confidentiality and
the device availability
MITM Attack: Listening to the Loss of privacy and data
Consumer and communication between two confidentiality
Home endpoints
Skill Squatting Attack: Exploiting Loss of privacy and data
the voice-based commands in digital confidentiality
assistants such as Alexa and Google
Home
Formjacking Attack: Stealing Loss of privacy and data
credit card details and personal
information from payment forms
Signal-Jamming Attack: Loss of data availability
Electromagnetic interference or
interdiction using the same
frequency-band wireless systems
Access Control: Gaining physical or Loss of confidentiality and
remote access to the device availability
DoS Attack: Making service Loss of data availability
Healthcare and Life
unavailable for legitimate users by
Science
flooding the system with
communication requests
Eavesdropping: Collecting Loss of data confidentiality
exchanged messages
Sinkhole Attack: Compromised Loss of data availability
nodes try to attract traffic by
advertising a fake route
Sybil Attack: The reputation Loss of data confidentiality
system is subverted by forging
multiple identities
Bluesnarfing Attack: Gaining Loss of privacy and
illegal access to Bluetooth devices to confidentiality
retrieve information
ZigBee End-Device (ZED) Loss of data availability
Sabotage Attack: Damages the
ZED by sending a signal periodically
to wake up the object to drain its
battery
MITM Attack: Listening to the Loss of privacy and data
communication between two confidentiality
endpoints
Impersonation Attack: Attacker Loss of privacy and data
successfully assumes the identity of confidentiality
the other legitimate user
Sybil Attack: The reputation Loss of data confidentiality
system is subverted by forging
multiple identities
GPS Spoofing: Deceiving a GPS Loss of data availability
receiver by broadcasting incorrect
GPS signals
DoS Attack: Making service Loss of data availability
unavailable for legitimate users by
flooding the system with
Transportation communication requests
/Automobile / Eavesdropping: Collecting Loss of data confidentiality
Security and Public exchanged messages
Safety Access Control: Gaining access to Loss of confidentiality and
the device availability
Wormhole Attack: Captures Loss of confidentiality and
packets from one location and sends availability
it to another network
Black Hole Attack: Router discards Loss of data
packets instead of relaying them
Bluesnarfing Attack: Gaining Loss of privacy and
access to Bluetooth devices illegally confidentiality
to retrieve information
ZigBee End-Device (ZED) Loss of data availability
Sabotage Attack: Damages the
ZED by sending a signal periodically
to wake up the object to drain its
battery
Brute Force: Generate many Loss of privacy and data
guesses to find the correct confidentiality
credentials to gain access to the
system
DoS Attack: Making service Loss of data availability
unavailable for legitimate users by
flooding the system with
communication requests
IT and Networks
Access Control: Gaining access to Loss of confidentiality and
the device availability
Ohigashi–Morii Attack: Attacks Loss of confidentiality and
the WPA-TKIP by reducing the availability
injection time of a malicious packet
SSL Stripping: Manipulates Loss of privacy and data
unencrypted protocols to demand confidentiality
the use of TLS
Jamming Attack: Prevents other Loss of data availability
nodes from using the channel to
communicate by occupying the
channel
Fragmentation Attack: Guess the Loss of privacy and data
Critical Water first 8 bytes of the headers by confidentiality
Infrastructure XORing

DoS Attack: Making service Loss of data availability


unavailable for legitimate users by
flooding the system with
communication requests
Exploitation of Loss of confidentiality and
Misconfiguration: Improper data availability
configuration of sensors or IoT
devices leading to exploitation of the
security device
Agriculture Path-Based DoS Attack: Injects Loss of data availability
malicious code into the packets or
replays some packets to the network
Reprogram Attack: Loss of privacy and data
Reprogramming the IoT device confidentiality
remotely
GPS Spoofing: Deceiving a GPS Loss of data availability
receiver by broadcasting incorrect
GPS signals
Signal-Jamming Attack: Loss of data availability
Electromagnetic interference or
interdiction using the same
Marine frequency-band wireless systems
Access Control: Gaining access to Loss of confidentiality and
the device availability
Redirecting Communication: Loss of privacy and data
Redirect and eavesdrop on the confidentiality
packets to intercept and change the
transmitted data
Table 18-04: IoT Application Areas and Attacks

IoT Malware
KmsdBot
KmsdBot is a malware specifically designed to target IoT devices and has evolved significantly
since its initial discovery. The most recent version, Kmsdx, adds new features, including telnet
scanning and authentication for legitimate telnet services. This updated malware scans random
IP addresses for open SSH ports and tries to access them using a password list obtained from
the Command-and-Control (C2) server. Additionally, the telnet scanner verifies valid services
on port 23, broadening the botnet's ability to target a larger variety of IoT devices.

Figure 18-17: KmsdBot Showing Added Code to Handle Telnet Scanning


The latest malware version supports various CPU architectures commonly used in IoT devices.
By incorporating telnet and SSH scanning, the new variant allows malware operators to exploit
IoT device vulnerabilities more efficiently. Additionally, the malware’s advanced legitimacy
checks, including the verification of data buffers, demonstrate the sophistication of its targeting
methods.
The KmsdBot scanning capability is particularly effective at identifying IoT devices that still use
default credentials, which users often leave unchanged. By exploiting default credentials found
in files like telnet.txt, which contains commonly used weak passwords, the malware’s impact is
significantly amplified, increasing the likelihood of IoT devices being compromised and added
to botnets.

Figure 18-18: Partial List of Downloaded Credentials Stored in Telnet.txt

Additional IoT malware:


● WailingCrab
● P2PInfect
● NKAbuse
● IoTroop
● XorDdos

Case Study: IZ1H9


IZ1H9, a botnet malware based on Mirai, was first identified in early 2021 and continued to
spread through 2023. The rise of botnet malware can be attributed to exploiting various
vulnerabilities in IoT networks. After identifying and infecting a vulnerable device, IZ1H9
incorporates it into the botnet network, taking control of its computational resources to
conduct Distributed Denial-of-Service (DDoS) attacks. The malware employs an advanced shell
script downloader to evade security measures, conceal its presence, and maintain a persistent
connection to a Command-and-Control (C2) server.

Figure 18-19: Illustration of IZ1H9 Infection Flow

IZ1H9 Attack Scenario:


● Step 1: Initial Exploitation
Attackers target devices running Linux operating systems, seeking out those with weak
configurations. They then exploit vulnerabilities in exposed servers and network devices within
the IoT environment. Their focus is on scanning and exploiting devices with the following
vulnerabilities:
o Tenda G103 command injection vulnerability
o LB-Link command injection vulnerability
o DCN DCBI-Netlog-LAB remote code execution vulnerability
o Zyxel remote code execution vulnerability
● Step 2: Exploitation
The attackers attempt to download and execute a shell script downloader (lb.sh) from the IP
address 163.123.143[.]126 on the identified vulnerable devices. This script deletes log files to cover
its tracks and then downloads bot clients for various Linux architectures, such as
o hxxp://163.123.143[.]126/bins/dark.x86
o hxxp://163.123.143[.]126/bins/dark.mips
o hxxp://163.123.143[.]126/bins/dark.mpsl
o hxxp://163.123.143[.]126/bins/dark.arm4
o hxxp://163.123.143[.]126/bins/dark.arm5
o hxxp://163.123.143[.]126/bins/dark.arm6
o hxxp://163.123.143[.]126/bins/dark.arm7
o hxxp://163.123.143[.]126/bins/dark.ppc
o hxxp://163.123.143[.]126/bins/dark.m68k
o hxxp://163.123.143[.]126/bins/dark.sh4
o hxxp://163.123.143[.]126/bins/dark.86_64

Both bot clients are fetched from URLs such as hxxp://2.56.59[.]215/i.sh and
hxxp://212.192.241[.]72/lolol.sh, which connect to a Command-and-Control (C2) domain,
dotheneedfull[.]club, resolving to the IP address 212.192.241.72. The botnet client ensures that
only one instance of the process runs. Suppose a botnet process is already active on the device.
In that case, the new client overwrites the existing process and initiates a fresh one. Then, the
bot clients set up an encrypted string table and use an index to retrieve and decrypt the strings.

Figure 18-20: The Initialization of the Encrypted String Table

Figure 18-21: IZ1H9 Retrieving Strings

The IZ1H9 variant decrypts its configuration strings by applying XOR decryption with the key
0xBAADF00D.
The IZ1H9 variant propagates via HTTP, SSH, and telnet channels. For SSH and telnet, it uses
brute-force attacks with a list of approximately 100 weak username and password pairs. For
HTTP, the variant exploits four remote code execution vulnerabilities to run shellcode scripts,
further compromising the devices.

Figure 18-22: Exploit Scanner Function

In the Tenda vulnerability exploitation function, the payload retrieves the file tenda.sh but runs
the script netlog.sh instead.

Figure 18-23: Malware Executes Wrong Downloader

● Step 3: Persistence
The botnet clients create persistent connections with hardcoded Command-and-Control (C2)
networks. Subsequently, the attacker configures a set of attack strategies for the targets,
utilizing specific command codes as detailed in Table 18-05.
Command Attack Method Description

0 attack_method_tcpsyn TCP SYN flooding attack


1 attack_method_tcpack TCP ACK flooding attack
2 attack_method_tcpusyn TCP URG-SYN flooding attack
3 attack_method_tcpall TCP DDoS with all options set
4 attack_method_tcpfrag TCP fragmentation attack
5 attack_method_asyn TCP SYN-ACK flooding attack
6 attack_method_udpgame UDP attack targets online gaming servers
7 attack_method_udpplain UDP flooding with fewer options
8 attack_method_greip GRE IP flooding attack
9 attack_method_std STD flooding attack
10 attack_method_udpdns DNS flooding attack
11 attack_method_udpgeneric UDP flooding attack
12 attack_app_http HTTP flooding attack
13 attack_method_dnsamp DNS amplification attack
Table 18-05: Commands for Various Attack Methods to be Initiated on Target IoT Network

EXAM TIP: Familiarize yourself with the fundamental principles of IoT. Focus on
understanding how these systems work, their architecture, and the main characteristics of IoT
devices. This foundational knowledge will help you tackle more complex questions.

IoT Hacking Methodology


Using the IoT hacking methodology, cyber attackers gather intelligence through information
collection, attack surface assessment, and vulnerability scans. They leverage this information to
compromise targeted devices and networks. This section delves into attackers' tools and
strategies to breach IoT devices.
What is IoT Device Hacking?
With the rapid expansion of IoT technology, an ever-growing number of devices are becoming
integral to our daily lives, from home automation systems to healthcare solutions. While IoT
devices bring convenience and efficiency, they also introduce significant cybersecurity risks.
Many IoT devices lack fundamental security measures, making them vulnerable to cyber
threats. Attackers exploit these vulnerabilities to gain unauthorized access to devices and
sensitive user data. Compromised IoT devices can be manipulated to create botnets, which
attackers often use to orchestrate large-scale Distributed Denial-of-Service (DDoS) attacks.
How can a hacker gain profit from the IoT when it is successfully compromised?
Today, smart devices and IoT devices hold vast amounts of personal data, including your
location, email accounts, financial details, and photos, making them prime targets for hackers.
With the growing popularity and widespread adoption of IoT devices, their numbers surpass
the global population. They are projected to reach 75 billion by 2025. However, the absence of
robust security measures makes these devices vulnerable to exploitation. Hackers can
compromise smart devices to monitor user behavior, misuse sensitive data (such as medical
records), deploy ransomware to lock users out of their devices, spy on individuals through CCTV
cameras, commit credit card fraud, access homes, or integrate devices into botnets for large-
scale DDoS attacks.
IoT Hacking Methodology
The process of hacking IoT devices typically involves the following phases:
● Information Gathering
● Vulnerability Scanning
● Launching Attacks
● Gaining Remote Access
● Maintaining Access
Information Gathering
The initial step in hacking an IoT device involves collecting critical information about the target.
This includes details such as the device’s IP address, communication protocols (e.g., Zigbee,
BLE, 5G, IPv6LoWPAN), open ports, device type, geographic location, serial number, and
manufacturer. During this phase, attackers also analyze the device's hardware design,
infrastructure, and key embedded components accessible online. Tools like Shodan, Censys,
and FOFA are commonly utilized to gather this information or conduct reconnaissance on the
target. Even devices not actively connected to a network but within the communication range
can be identified using sniffing tools such as Suphacap, CloudShark, and Wireshark.
Information Gathering Using Shodan
Shodan is a specialized search engine designed to index and provide information on internet
connected devices, including routers, traffic lights, CCTV cameras, servers, smart home gadgets,
and industrial equipment. Hackers can exploit this tool to extract details like IP addresses,
hostnames, ISPs, device locations, and banners associated with IoT devices. Shodan offers filters
to refine searches and gather specific data on target devices. Examples include:
● Search by Geolocation:
webcamxp country: US (Retrieves all webcamxp webcams in the United States)
● Search by City:
webcamxp city:paris (Lists webcamxp webcams in Paris)
● Search by Coordinates:
webcamxp geo:-50.81,201.80 (Identifies webcams at the coordinates "-50.81,201.80," such as in
Boston, USA)
Other filters commonly used for targeting include:
● Net: Filter devices by IP address or CIDR range
● OS: Search for devices based on their operating systems
● Port: Identify open ports
● Before/After: Narrow results to a specific timeframe
Figure 18-24: Information Gathering Using Shodan

Attackers can utilize the MultiPing tool to identify the IP address of IoT devices within a
targeted network. Once the IP address is obtained, further scans can be conducted to uncover
vulnerabilities in the device. The steps to identify an IoT device's IP address using MultiPing are
as follows:
1. Launch the MultiPing application and navigate to File → Add Address Range.
2. In the Add Range of Addresses pop-up window:
o Select the router’s gateway IP address from the Initial Address to Add dropdown.
o Set the Number of addresses to 255.
o Click OK.

Figure 18-25: Adding a Range of IP Addresses in MultiPing

3. MultiPing will scan the specified address range, testing each IP address that responds to a
ping.
4. Each row in the MultiPing interface represents a device on the network. The attacker can
identify the IP address of the targeted IoT device from this list.
5. To expedite the process, adjust the ping interval to 1 second.

Figure 18-26: Scanning Using MultiPing

Information Gathering Using FCC ID Search


The FCC ID Search tool enables users to access detailed information about devices and their
certifications. Each device is uniquely labeled with an FCC ID, which comprises two parts: the
grantee ID (the first three or five characters) and the product ID (the remaining characters).
This ID allows users to retrieve information about a target device by following these steps:
1. Locate and examine the label attached to the device containing its FCC ID.
Figure 18-27: FCC ID location

2. Visit the official FCC ID search page at https://www.fcc.gov/oet/ea/fccid.


3. Enter the grantee code and product ID in their respective fields.

Figure 18-28: FCC ID Search Form

4. Click Search to display details and frequency summaries related to the device.
Figure 18-29: FCC ID Search Result

o To view the basic device details, click on the Summary link.

Figure 18-30: Summary Details of a Device

o For additional information, click the Detail link, which provides access to
documents such as cover letters, external and internal photos, test reports, and user
manuals.
Figure 18-31: Complete Details of a Device

Using this data, attackers can identify vulnerabilities in the target device and potentially exploit
them to carry out further attacks.
Information-Gathering Tools
Hackers employ tools like Shodan and Censys to collect fundamental details about a target
device and its network. These tools provide insights into live devices on the network, their
manufacturers, open ports, running services, and physical locations.
Censys
● Website: https://censys.io
Censys is a public search engine and data analysis tool powered by comprehensive internet wide
scanning. It offers full-text searches on protocol banners and access to various derived fields,
enabling the identification of vulnerable devices and networks. Censys provides real-time
monitoring of servers and devices on the internet, allowing security professionals to analyze
them as they change. Features include:
o Identifying specific vulnerabilities
o Generating statistical reports on device usage and trends
o Assessing network attack surfaces, discovering threats, and evaluating their global
impact
Censys compiles its data by performing daily IPv4 scans using tools like ZMap and ZGrab,
maintaining a detailed database of device and website configurations.
Figure 18-32: Censys

FOFA
● Website: https://en.fofa.info
FOFA is a cyberspace mapping and intelligence platform that enables data collection on IoT
devices worldwide. It helps attackers identify external attack surfaces, uncover internet-
connected resources, evaluate risks associated with exposed assets, and gather intelligence.
With FOFA, attackers can analyze profiles and vulnerabilities related to open internet assets
from any global location.
Figure 18-33: FOFA

Information Gathering through Sniffing


Many IoT devices, like security cameras, offer web-based interfaces for remote configuration
and control. These interfaces often rely on the unsecured HTTP protocol instead of HTTPS,
making them susceptible to cyberattacks. Suppose default factory credentials are still in use. In
that case, attackers can intercept traffic between the camera and its web application, potentially
gaining unauthorized access to the camera. By utilizing tools like Wireshark, attackers can
capture and decrypt the Wi-Fi keys of the target network.
Steps Attackers Use to Capture Wireless Traffic of a Web Camera
1. Identify IoT Devices with Insecure HTTP Ports:
Use Nmap to scan for devices transmitting data over unsecured HTTP ports:
nmap -p 80,81,8080,8081 <Target IP address range>
2. Set Wireless Card to Monitor Mode:
● Run ifconfig to find your wireless card (e.g., wlan0)
● Activate monitor mode using:

airmon-ng start wlan0


3. Scan Wireless Networks:
Use Airodump-ng to discover nearby networks:
airodump-ng start wlan0mon
Identify the target network and note its channel.
4. Listen to Traffic on the Target Channel:
Configure your wireless card to monitor the target network’s channel (e.g., channel 11):
airmon-ng start wlan0mon 11
5. Capture Traffic with Wireshark:
Open Wireshark, select the interface in monitor mode (e.g., wlan0mon), and begin capturing
traffic.
After capturing the data, attackers may decrypt the WEP or WPA keys using Wireshark. This
allows them to compromise the IoT device and access sensitive information.

Figure 18-34: Wireshark

Sniffing using Cascoda Packet Sniffer


Attackers may exploit the Cascoda Packet Sniffer to intercept IEEE 802.15.4 traffic from IoT
protocols like Thread, Zigbee, and KNX-IoT. This tool employs the Chili2D packet sniffing
firmware to capture traffic passively. It then analyzes and decrypts the packet headers using
Wireshark, enabling real-time packet monitoring or saving the data in PCAP format.
Additionally, the tool provides metrics such as signal strength and link quality for the sniffed
IoT traffic.
Steps for Capturing and Analyzing IoT Traffic with Cascoda Packet Sniffer
1. Download and Configure Tools:
The attacker installs Cascoda’s tools on a Windows machine and configures Wireshark to
monitor network traffic.
2. Connect the Sniffer Device:
The Cascoda Packet Sniffer dongle is plugged into the attacker’s system.

Figure 18-35: Cascoda Packet Sniffer

3. Start Packet Capture:


The attacker initiates live packet capture on a specific channel by executing the command:
sniffer -w <channel_number>
4. Analyze Traffic in Wireshark:
Wireshark captures the IoT network traffic in real-time. The attacker applies relevant display
filters to isolate and analyze specific data packets.

Figure 18-36: Wireshark Displaying IoT Traffic Sniffed Using Cascoda Packet Sniffer
This process enables attackers to gain insights into IoT communications, potentially uncovering
vulnerabilities or sensitive data.
Sniffing Tools
Network administrators rely on automated tools to oversee their systems and devices, ensuring
smooth operations. However, attackers often exploit these same tools to intercept and analyze
network data for malicious purposes. Below are examples of tools that attackers might use to
capture traffic from IoT devices:
● Suphacap
Suphacap is a hardware-based Z-Wave traffic sniffer designed to monitor and capture data
packets from smart devices within a network. This tool provides real-time packet capture across
all Z-Wave networks. It is compatible with various Z-Wave controllers, such as Fibaro,
Homeseer, Tridium Niagara, Z-Way, SmartThings, and Vera.

Figure 18-37: Wave Sniffer

Figure 18-38: Z-Wave Sniffer Displaying Raw Traffic

Other Tools for Sniffing IoT Device Traffic


Attackers also leverage additional tools to intercept and analyze traffic from IoT networks.
These tools enable the monitoring of communications, revealing potentially exploitable
vulnerabilities.
● IoT Inspector 2 (https://github.com)
● ZBOSS Sniffer (https://dsr-iot.com)
● tcpdump (https://www.tcpdump.org)
● Ubiqua Protocol Analyzer (https://www.ubilogix.com)
● Perytons Protocol Analyzers (https://www.perytons.com)
Vulnerability Scanning
After collecting information about a target device, attackers identify its potential points of
attack, known as attack surfaces, where vulnerabilities may exist. Vulnerability scanning enables
attackers to uncover weaknesses in the device's exposed firmware, infrastructure, and system
components. Once the attack surface is identified, attackers perform scans to detect specific
vulnerabilities, which could serve as pathways for further exploitation. This scanning process
helps attackers pinpoint IoT devices with weak configurations, such as hidden flaws, firmware
defects, insecure settings, weak passwords, and insufficient encryption. On the flip side,
vulnerability scanning is also a valuable tool for security experts, as it allows them to identify
security gaps in IoT devices before attackers can exploit them, thereby strengthening network
defenses.
Vulnerability Scanning Using IoTSeeker
Attackers utilize tools like IoTSeeker to identify IoT devices that are still using default
credentials and are susceptible to hijacking attacks. IoTSeeker scans networks for specific IoT
devices and checks whether they operate with factory-set credentials. A recent internet outage
was linked to IoT devices such as CCTV cameras and DVRs still using default settings. This tool
assists organizations in scanning their networks to detect such devices, helping them determine
whether the credentials have been altered or are still operating with the default configuration.
IoTSeeker primarily focuses on HTTP/HTTPS services. For instance, attackers may execute the
following command to search for devices with default credentials:
perl iotScanner.pl <IP address/range of IPs>

Figure 18-39: IoTSeeker


Vulnerability Scanning Using Genzai
Genzai is an IoT security toolkit attackers use to detect and scan IoT dashboards, such as
wireless routers, surveillance cameras, and Human-Machine Interfaces (HMIs), for default
passwords and vulnerabilities related to paths and versions. This tool identifies the IoT product
running on a target by analyzing HTTP responses using predefined signatures and templates. It
then searches for vendor-specific default credentials like 'admin:admin' and other potential
security weaknesses. To scan a target IoT device’s dashboard and save the output in .json format,
attackers can use the following command:
./genzai <target_host> -save scan.json

Figure 18-40: Genzai

Vulnerability Scanning using Nmap


Attackers employ vulnerability-scanning tools like Nmap to detect IoT devices on a network,
including their open ports and services. Nmap sends raw IP packets in various formats to
identify active devices, the services they provide, their operating systems, and the packet filters
in use. Another tool commonly used for this purpose is an Angry IP Scanner. The following
Nmap command is used to scan a specific IP address:
nmap -n -Pn -sS -pT:0-65535 -v -A -oX <Name><IP>
To perform a comprehensive scan that includes both TCP and UDP services and ports, attackers
can use:
nmap -n -Pn -sSU -pT:0-65535,U:0-65535 -v -A -oX <Name><IP>
For identifying the IPv6 capabilities of a device, the following Nmap command is used:
nmap -6 -n -Pn -sSU -pT:0-65535,U:0-65535 -v -A -oX <Name><IP>

Figure 18-41: Scanning using Nmap

Vulnerability Scanning Tools


Vulnerability scanning helps attackers identify security flaws in IoT devices and their networks,
enabling them to determine potential exploitation methods. These tools also benefit network
security professionals, providing suggestions for mitigating weaknesses and safeguarding the
network.
● beSTORM
beSTORM is an automated fuzzing tool designed to detect buffer overflow vulnerabilities by
generating corrupted inputs and monitoring the system for unexpected reactions. Using
protocol-based fuzzing, it serves as an automated black-box testing tool, testing numerous
attack scenarios, starting with the most likely ones, to uncover application anomalies that signal
successful attacks. It helps identify coding flaws and evaluates the security resilience of any
product, even without access to the source code. beSTORM supports testing various protocols
and hardware, including those used in IoT, automotive, process control, and aerospace systems.
Figure 18-42: beSTORM

Additional vulnerability scanners for IoT devices are listed below:


● Metasploit (https://www.rapid7.com)
● IoTsploit (https://iotsploit.co)
● IoTSeeker (https://www.rapid7.com)
● IoTVAS (https://firmalyzer.com)
● Enterprise IoT Security (https://www.paloaltonetworks.com)
Analyzing Spectrum and IoT Traffic
Spectrum Analysis With Gqrx
Gqrx is a Software-Defined Radio (SDR) that integrates with GNU Radio and the Qt GUI tool.
Attackers often use Gqrx with hardware devices like the FunCube Dongle, Airspy, HackRF, RTL-
SDR, and USRP to analyze the radio spectrum. This setup allows them to monitor frequency
bands used by devices such as temperature/humidity sensors, light switches, car key fobs, and
M-bus transmitters. Gqrx also enables attackers to listen to FM radio frequencies or other radio
communications.
Steps to Analyze Spectrum with Gqrx:
1. Install Gqrx and GNU Radio packages by running the following commands to download
from GitHub:

git clone https://github.com/gqrx-sdr/gqrx.git gqrx.git


cd gqrx.git
mkdir build
cd build
cmake ..
make
Hardware tools, like the FunCube Dongle Pro+, are connected to a PC's USB port to analyze
various frequency ranges.
2. Launch Gqrx using:

gqrx
3. The main Gqrx window will display frequencies, and their sounds can be monitored through
speakers or headphones.

Figure 18-43: Gqrx

Attempters can adjust the FFT settings (located in the bottom-right corner) to capture and
analyze different nearby frequencies.
IoT Traffic Analysis with ONEKEY
ONEKEY is a tool used by attackers to discover IoT devices on a network and analyze their
traffic to find security vulnerabilities. This tool, along with the IoT Inspector, helps attackers
bypass privacy and security measures. ONEKEY presents vulnerabilities through tables and
graphs, making it easier for attackers to understand potential risks. It also allows for recording
and replaying data exchanged between IoT devices, helping capture sensitive information.
ONEKEY scans the network automatically to identify available devices. After selecting a target
device, attackers can view its network activity and communication endpoints. By clicking
"network activities," a live chart shows the traffic the device is handling, while the
communication endpoints display the services the IoT device interacts with.

Figure 18-44: ONEKEY

Expanded version of all labs with screenshots: https://github.com/IP-Specialist/CEHv13---


Hands-on-Labs/blob/main/Topic%2018%20IoT%20and%20OT%20Hacking.zip

Tools to Perform SDR-Based Attacks


Attackers utilize various tools, including RTL-SDR, GNU Radio, and Universal Radio Hacker
(URH), to carry out different types of attacks, such as reconnaissance, replay, and cryptanalysis,
on devices that use Software-Defined Radio (SDR).
Universal Radio Hacker
Universal Radio Hacker (URH) is a tool that analyzes unknown wireless protocols used by
various IoT devices. It allows attackers to:
o Identify hardware interfaces for common SDRs
o Demodulate signals
o Manage data flow by assigning roles to participants
o Decode advanced encodings like CC1101 data whitening
o Label data to understand protocol logic
o Automate reverse engineering of protocol fields
o Use fuzzing techniques to detect security vulnerabilities
o Perform modulation to inject data back into the system
Figure 18-45: Universal Radio Hacker

Additionally, there are other tools available for conducting SDR-based attacks.
● BladeRF (https://www.nuand.com)
● TempestSDR (https://github.com)
● HackRF One (https://greatscottgadgets.com)
● GP-Simulator (https://gpspatron.com)
● Gqrx (https://gqrx.dk)
Launch Attacks
During the vulnerability scanning phase, attackers attempt to identify weaknesses in the target
device. Once these vulnerabilities are discovered, they can be exploited to initiate various
attacks, such as DDoS, rolling-code, signal jamming, Sybil, MITM, and data or identity theft.
For instance, an attacker might use the RFCrack tool to execute attacks like rolling-code, replay,
and jamming. Similarly, tools like KillerBee can target ZigBee and IEEE 802.15.4 networks.
Rolling Code Attack Using RFCrack
Attackers use the RFCrack tool to intercept and capture the rolling code transmitted by a victim
to unlock a vehicle, allowing them to replay the code later and steal it. RFCrack is designed to
test RF communications between devices operating on sub-GHz frequencies. It works with
hardware such as yardsticks to jam, replay, and sniff the signals the device sends. Attackers can
use RFCrack to execute various types of attacks, including:
● Replay attacks (-i -F)
● Sending previously captured payloads (-s -u)
● Rolling code bypass attacks (-r -F -M)
● Jamming (-j -F)
● Scanning frequencies incrementally (-b -v -F)
● Scanning common frequencies (-k)
Some of the specific commands used by attackers to perform a rolling code attack include:
 Live replay:

python RFCrack.py -i
 Rolling code:

python RFCrack.py -r -M MOD_2FSK -F 314350000


 Adjust RSSI range:

python RFCrack.py -r -M MOD_2FSK -F 314350000 -U -100 -L -10


 Jamming:

python RFCrack.py -j -F 314000000


 Scan common frequencies:

python RFCrack.py -k
 Scan with a custom list:

python RFCrack.py -k -f 433000000 314000000 390000000


 Incremental scan:

python RFCrack.py -b -v 5000000


 Send saved payload:

python RFCrack.py -s -u ./captures/test.cap -F 315000000 -M MOD_ASK_OOK


Figure 18-46: Rolling Code Attack Using RFCrack

Hacking Zigbee Devices with Open Sniffer


Open Sniffer is a Wireshark-based tool attackers use to capture and analyze network traffic from
networks like IEEE 802.15.4, Zigbee, 6LoWPAN, Wireless Hart, and ISA100.11a. It uses two
multiband antennas and USB-powered cables to connect to the target device or network switch
through an Ethernet port. This tool allows attackers to capture all 802.15.4 frames within its
range and decode them, displaying the frames on Wireshark for further analysis. Attackers can
also send packets to a selected channel with specific transmission types, modulation, and power
levels to launch a Denial-of-Service (DoS) attack. Additionally, Open Sniffer can replay captured
frames to perform replay attacks on the target network or device. It also provides energy
detection across all channels, displaying the results graphically.
Figure 18-47: Open Sniffer

Figure 18-48: Wireshark Displaying the Packets Captured by Open Sniffer

BlueBorne Attack Using HackRF One


IoT devices rely on wireless communication technologies like RF, ZigBee, or LoRa. HackRF One
is a device commonly used by attackers to execute various attacks, such as BlueBorne or
AirBorne attacks, along with replay, fuzzing, and jamming. This advanced software-defined
radio device operates in the frequency range of 1 MHz to 6 GHz. It works in half-duplex mode,
allowing attackers to transmit and receive radio signals easily. HackRF One can intercept a
broad spectrum of wireless protocols, ranging from GSM to Z-Wave, making it a versatile tool
for cyber attackers.
Figure 18-49: HackRF One

Replay Attack Using HackRF One


Attackers use tools like HackRF One to perform replay attacks on IoT devices. The first step in
this process involves discovering the target device's radio frequency. To find this frequency,
attackers may consult online resources like the FCC database or use tools such as RTL-SDR to
detect the frequency of nearby devices. Once the target frequency is identified, attackers can
proceed with the replay attack using HackRF One.
The steps for executing a replay attack on an IoT device are as follows:
 Step 1: Capture the device's signal using the command:

hackrf_transfer -r connector.raw -f [device frequency]


Here, -r indicates recording, and -f specifies the device's frequency.

Figure 18-50: HackRF One Recording Signal

 Step 2: Replay the captured signal to the target device using the command:

hackrf_transfer -t connector.raw -f [device frequency]


Here, -t is used to replay the signal.
Figure 18-51: HackRF One Replaying a Signal

Once the replay attack is successful, the attacker can take control of the IoT device to carry out
additional malicious activities.
SDR-Based Attacks using RTL-SDR and GNU Radio
Hardware-based Attack
Attackers utilize hardware tools like RTL-SDR to carry out SDR-based attacks on IoT devices.
o RTL-SDR
RTL-SDR is a hardware device available as a USB dongle. It enables attackers to capture radio
signals in the surrounding area without needing an internet connection. It comes in various
models, including DVB-T SDR, RTL2832, RTL dongle, and DVB-T dongle. Depending on the
chosen SDR model, RTL-SDR can capture frequencies from 500 kHz up to 1.75 GHz.

Figure 18-52: RTL-SDR


Using an RTL-SDR radio scanner, attackers can perform several activities, including:
 Receiving and decoding GPS signals
 Analyzing the frequency spectrum
 Listening to DAB broadcast radio
 Decoding HD radio
 Sniffing GSM signals
 Monitoring VHF amateur radio
 Scanning trunked radio conversations
 Searching for cordless phone communications
Software-Based Attack
In addition to hardware tools, attackers can also use various software tools, such as GNU Radio,
to target SDR-based IoT devices.
o GNU Radio
GNU Radio is a tool that uses external RF hardware to generate Software-Defined Radio (SDR)
signals. It provides the framework and tools to generate SDR signals and includes signal
processing units to implement software radios. Attackers use GNU Radio to carry out SDR-
based attacks on IoT devices. Before launching an attack, attackers must first build and
configure GNU Radio. Once GNU Radio is successfully installed, attackers can utilize the
following tools for exploitation:
● uhd_ft: A spectrum analyzer that connects to a UHD device to scan the spectrum at a
specific frequency
● uhd_rx_cfile: Stores wave samples using a UHD device, which can later be analyzed using
GNU Radio or tools like Matlab or Octave
● uhd_rx_nogui: Allows attackers to capture and listen to incoming signals on an audio
device
● uhd_siggen_gui: A tool to generate signals such as sine, square, or noise
● gr_plot: Used to display recorded samples stored in a file
Figure 18-53: GNU-Radio

Side-Channel Attack Using ChipWhisperer


ChipWhisperer is an open-source toolchain designed primarily for embedded hardware security
research. It focuses on side-channel power analysis and glitching attacks, which are typically
used to extract cryptographic keys from systems. A side-channel attack targets the physical
system's implementation to gather sensitive information, such as power consumption, timing,
sound, or electromagnetic emissions, rather than exploiting software vulnerabilities.
To conduct a side-channel attack using ChipWhisperer, two key components are required:
 Capture Board: This specialized hardware captures minuscule signals synchronized with a
precise clock.
 Target Board: A programmable processor used to perform secure operations.
Attackers leverage ChipWhisperer to break down complex cryptographic algorithms, such as
AES and Triple DES, using power analysis attacks, which fall under the category of side-channel
attacks. This technique involves controlling the input data and power consumption. The known
input is XORed with the unknown, producing unknown output data. The attacker then
compares a guessed secret key with real measurements to reveal the original key.
Side-channel attacks include cache, timing, power-monitoring, electromagnetic, acoustic
cryptanalysis, fault analysis, data remanence, and optical attacks. ChipWhisperer can also inject
glitches into embedded hardware, allowing attackers to manipulate the system's behavior by
altering either the clock or input power, potentially disclosing sensitive information.
Figure 18-54: ChipWhisperer

Identifying IoT Communication Buses and Interfaces


Attackers look for various serial and parallel interfaces like Universal Asynchronous Receiver-
Transmitter (UART), Serial Peripheral Interface (SPI), Joint Test Action Group (JTAG), and
Inter-Integrated Circuit (I2C) to gain unauthorized access to devices, extract firmware, and
perform other malicious activities. Tools such as BUS Auditor, Damn Insecure and Vulnerable
Application (DIVA), and Printed Circuit Boards (PCBs) are used to identify these interfaces. BUS
Auditor features 16 independent channels (CH0 to CH15). The ground pins of the BUS Auditor
and the DIVA IoT board are connected to begin the process.

Figure 18-55: Bus Auditor

UART
The steps to identify UART on a PCB without microcontroller data are as follows:
1. Connect channels CH0 and CH1 of the BUS Auditor to the UART header.
2. Connect the DIVA IoT board and the BUS Auditor to the computer.
3. Execute the following command in the EXPLIoT framework:

run busauditor.generic.uartscan -v 3.3 -p /dev/ttyACM0 -s 0 -e 1 -v


Commands Description
-v Voltage
-p Device Port
-s Starting Channel
-e Ending Channel
Table 18-06: Command Description

Figure 18-56: UART Scan

JTAG
Joint Test Action Group (JTAG), standardized as IEEE 1149.1, consists of four primary pins—Test
Mode Select (TMS), Test Clock (TCK), Test Data In (TDI), and Test Data Out (TDO)—with an
optional fifth pin, Test Reset (TRST).
To identify JTAG, follow these steps:
1. Connect channels CH0 to CH8 of the BUS Auditor to the JTAG header.
2. Connect the DIVA board and the BUS Auditor to the computer.
3. Execute the following command in the EXPLIoT framework:

run busauditor.generic.jtagscan -v 3.3 -p /dev/ttyACM0 -s 0 -e 10


Figure 18-57: JTAG Scan

I2C
Inter-Integrated Circuit (I2C) uses two main lines: Serial Data (SDA) for data transmission and
reception and Serial Clock (SCL) for synchronization.
To identify I2C, follow these steps:
1. Connect channels CH0 to CH8 of the BUS Auditor to the I2C header.
2. Connect the DIVA board and the BUS Auditor to the computer.
3. Run the following command in the EXPLIoT framework:

run busauditor.generic.i2scan -v 3.3 -p /dev/ttyACM0 -s 0 -e 10


Figure 18-58: I2C Scan

SPI
Attackers may use Google searches to identify the Serial Peripheral Interface (SPI) and its
pinouts by referencing chip numbers.

Figure 18-59: SPI Chip

Below are additional tools used for interface identification:


● JTAGulator (https://grandideastudio.com)
● Attify Badge (https://www.attify-store.com)
● Saleae Logic Analyzer (https://www.saleae.com)

NAND Glitching
Attackers frequently target IoT devices or routers to gain privileged access by exploiting boot
vulnerabilities, often using techniques like glitching. NAND glitching involves obtaining
privileged root access during the boot process by manipulating the serial I/O pin of a flash
memory chip. This method takes advantage of a vulnerability where a backup process in the
local flash memory grants single-user access in case of a boot failure. A precisely timed glitch,
lasting only about 1 ms, can lead to root access.
Steps for Performing NAND Glitching:
1. Reconnaissance:
Start by running the following command to capture the boot logs through a UART-USB
converter:
minicom -D /dev/ttyUSB0 -w -C D-link_startup.txt
This command helps gather the boot logs, which help identify the flash memory chip that holds
the boot firmware.
2. Glitching:
o Short the serial I/O pin of the flash memory chip to ground, disrupting the boot process and
triggering the backup loader code.

Figure 18-60: NAND Glitching


Figure 18-61: Interruption to Device Boot up

o Use the printenv command to view the boot arguments, which will show output like:

bootargs=noinitrd console=ttyAM0,115200 rootfstype=ubifs ubi.mtd=5 root=ubi0:rootfs


rw gpmi badupdater

Figure 18-62: Output of Bootlogs on the Console

o Set the environment variables using:

setenv bootargs 'noinitrd console=ttyAM0,115200 rootfstype=ubifs ubi.mtd=5


root=ubi0:rootfs rw gpmi init=/bin/sh'
o Finally, use the following command to load the backup privileged booting image from the
flash memory:
nand read ${loadaddr} app-kernel 0x00400000 && bootm ${loadaddr}
In this process, the bootm command loads the backup image, granting the attacker root access
to the device.

Figure 18-63: Root Access on the Device

Exploiting Cameras Using CamOver


CamOver is a tool attackers use to compromise network cameras and retrieve administrator
passwords. It targets vulnerabilities in well-known camera models like CCTV, GoAhead, and
Netwave. Additionally, the tool is designed to exploit multiple cameras simultaneously through
a threaded list.
Figure 18-64: CamOver

Here are some example commands for using CamOver:


● To start the tool, run the camover command in the terminal
● To exploit a specific camera with a given IP address, use the following command:

camover -a <Camera IP Address>


● To target cameras connected to the internet via the Shodan search engine, execute the
command:

camover -t --shodan <Shodan API Key>

Gain Remote Access


Vulnerabilities discovered during the scanning process enable attackers to remotely access and
control the attack while bypassing detection by security systems such as IDS/IPS, firewalls, and
antivirus software. Depending on the weaknesses in an IoT device, an attacker may convert the
device into a backdoor, allowing them to infiltrate an organization's network without
compromising any endpoints protected by security measures. Once remote access is gained,
attackers can use these devices as a launchpad to target other devices within the network.

Gaining Remote Access Using Telnet


Attackers use port scanning to detect open ports and services on a target IoT device. If they find
that the telnet port is open, they exploit this vulnerability to gain remote access. Many IoT
devices, including industrial control systems, routers, VoIP phones, and televisions, use telnet
to enable remote access, typically through a telnet server.
Once the attacker identifies an open telnet port, they can investigate the data exchanged
between devices, including details about their software and hardware models. The attacker then
looks for additional vulnerabilities to exploit. First, they check if authentication is required. If
not, they gain unauthorized access to explore the device’s stored data. If authentication is
needed, the attacker attempts default credentials, such as root/root or system/system, or uses
brute-force methods to crack the passwords for administrator or user accounts. Tools like
Shodan and Censys can help attackers gain remote access to the device.

Figure 18-65: Gaining Remote Access Using Shodan

Maintain Access
After gaining access to the device, attackers employ various methods to retain control and
further exploit. They may erase logs, update firmware, and deploy malicious programs like
backdoors and trojans to stay undetected. Additionally, attackers use tools such as Firmware
Mod Kit, IoTVAS, Firmware Analysis Toolkit, and Firmwalker to manipulate and exploit the
device’s firmware.
Maintain Access by Exploiting Firmware
The Firmware Mod Kit is a toolset designed to simplify the process of deconstructing and
reconstructing firmware images for various embedded devices. While it is mainly intended for
Linux-based routers, it is also compatible with most firmware that utilizes common formats and
file systems such as TRX/uImage and SquashFS/CramFS. This kit consists of tools, utilities, and
shell scripts, which can be used individually or in combination to automate common firmware
tasks (e.g., extraction and rebuilding). Using the Firmware Mod Kit, attackers can perform
actions such as:
● Extracting a firmware image into its components
● Modifying the firmware’s file system or web interface (webif) as desired
● Rebuilding the firmware
● Flashing the altered firmware onto the device and brick it
The core scripts to facilitate firmware operations are listed in Table 18-07.

Primary Scripts Secondary Scripts

extract-firmware.sh → Firmware extraction script ddwrt-gui-extract.sh → Extracts web GUI files from
extracted DD-WRT firmware
build-firmware.sh → Firmware rebuilding script ddwrt-gui-rebuild.sh → Restores modified web GUI
files to extracted DD-WRT firmware
Table 18-07: Firmware Mod Kit Code Scripts

Figure 18-66: Firmware Mod Kit

Firmware Analysis and Reverse Engineering


Firmware is crucial in controlling the functions of various IoT devices, and attackers often
analyze it to discover vulnerabilities and weaknesses. By inspecting the firmware, attackers can
identify sensitive information such as passwords, API tokens, vulnerable services, backdoor
accounts, configuration files, private keys, and stored data. Below are the steps attackers
typically follow for firmware analysis and reverse engineering:
1. Obtain Firmware
After gaining access to the target IoT device, attackers extract the firmware.
2. Analyze Firmware
To analyze the firmware, attackers use the following commands:
 Check the firmware file type:
Run the “file” command on the *.bin file.
 Verify MD5 Signature:
o Run the “cat” command on the *.md5 file
o Run the “md5sum” command on the *.bin file
 Extract Strings:
o Run “strings” against the *.bin file
Example:
strings -n 10 xyz.bin > strings.out
less strings.out
 Hexdump the File:
Run “hexdump” against the *.bin file
Example:
hexdump -C -n 512 xyz.bin > hexdump.out
cat hexdump.out
Running hexdump can help identify the type of firmware build
3. Extract the Filesystem
 Use Binwalk:
Run binwalk to analyze and reverse-engineer the firmware image, identifying the file system
type.
Example:
binwalk xyz.bin
 Extract the Filesystem with “dd”:
To extract the file system run:
dd if=xyz.bin bs=1 skip=922460 count=2522318 of=xyz.squashfs
4. Mount the Filesystem
 Create a mount directory, e.g., mkdir rootfs.
 Mount the filesystem with

sudo mount -t ext2 {filename} rootfs


4. Analyze Filesystem Content
Once the filesystem is mounted, check important files and directories:
etc/passwd, etc/shadow, etc/ssl
To search for sensitive data (e.g., password, admin, and root), use commands like
grep -rnw '/path/to/somewhere/' -e "pattern"
find . -name '*.conf' to locate configuration files or files such as *.pem, *.crt, *.cfg, .sh, and
.bin.
Tools like the Firmwalker script can also automate the search for these items within the
extracted filesystem.
5. Emulate Firmware for Dynamic Testing
For dynamic testing, use emulation software like QEMU or the Firmware Analysis Toolkit:
 Identify CPU Architecture:
Use commands like file or readelf to determine the CPU architecture of the firmware.
 Run User-mode Emulation:
For specific architectures, use commands such as:
qemu-mipsel -L <prefix> <binary>

qemu-arm -L <prefix> <binary>

qemu-<arch> -L <prefix> <binary>


Alternatively, use QEMU to perform cross-architecture emulation by transferring the qemu-
<arch>-static binary to the firmware’s root filesystem folder /usr/bin/ and running the
following:
chroot ~/<filename> /bin/

IoT Hacking Tools


Attackers utilize various IoT hacking tools to collect information about devices connected to a
network, including open ports, services, the extent of the attack surface, and existing
vulnerabilities. This information is then used to exploit individual devices and the broader
organizational network. Below are some IoT hacking tools commonly employed by attackers to
carry out attacks such as Distributed Denial-of-Service (DDoS), jamming, and BlueBorne
attacks.
CatSniffer
CatSniffer is a tool used by attackers to observe IoT network traffic passively. It helps gather
data about devices, communication protocols, and their exchanges. Attackers can target weak
points within the IoT network by identifying potential vulnerabilities. Network administrators
and security experts can also use CatSniffer to monitor and secure IoT environments, gaining
insights into the data shared across connected devices.
Figure 18-67: CatSniffer

Some additional tools to perform IoT hacking include:


● KillerBee (https://github.com)
● JTAGULATOR (https://www.grandideastudio.com)
● wiz_exploit (https://github.com)
● PENIOT (https://github.com)
● RouterSploit (https://github.com)

EXAM TIP: Consider the threats specific to IoT environments, such as DDoS attacks,
BlueBorne vulnerabilities, or rolling code attacks. Knowing these threats will enhance your
ability to identify them in exam scenarios.

IoT Attack Countermeasures


This section explores IoT security strategies, device management practices, and security tools
to safeguard IoT devices and networks. These measures aim to prevent, detect, and mitigate
different attacks while enabling recovery from potential breaches. Organizations can establish
robust security protocols by adopting these countermeasures to protect sensitive data
exchanged between IoT devices and the corporate network.
How to Defend Against IoT Hacking
1. Account Management:
o Disable unused accounts like "guest" and "demo"
o Use lockout features to prevent brute-force attacks after repeated failed login attempts
o Enforce strong authentication mechanisms
2. Network and Access Controls:
o Place IoT systems behind firewalls and isolate them from business networks
o Deploy Intrusion Detection System/Intrusion Prevention System (IDS/IPS)
o Restrict access to trusted IPs only
o Disable telnet (port 23) and UPnP on routers
3. Encryption and Secure Communication:
o Use end-to-end encryption and Public Key Infrastructure (PKI)
o Employ secure communication via VPNs
4. Device Hardening:
o Protect devices from physical tampering
o Regularly patch vulnerabilities and update firmware
o Enable secure boot to ensure only OEM-authorized code is executed
5. Authentication and Password Policies:
o Implement two-way authentication with cryptographic algorithms (e.g., SHA-HMAC,
ECDSA)
o Require strong passwords (8–10 characters with letters, numbers, and symbols)
o Use CAPTCHA and account lockout policies to prevent brute-force attacks
6. Data Security:
o Protect data confidentiality with symmetric encryption
o Ensure data authentication to verify the source
o Implement privacy measures to hide user identities
7. Traffic and Port Monitoring:
o Monitor port 48101 for malicious activity
o Use DNS filtering tools like dnswall to prevent DNS rebinding attacks
8. Configuration Management:
o Change router default settings (e.g., name and password)
o Disable unnecessary features on IoT devices
9. Advanced Security Measures:
o Secure keys and credentials using trusted modules like SAM, TPM, or HSM
o Validate code to prevent TOCTOU attacks
o Use trusted execution environments (e.g., ARM TrustZone)
10. Web and Cloud Security:
o Prevent IP address disclosure by disabling WebRTC in browsers
o Use ad-blockers and non-trackable extensions for web safety
o Leverage cloud-based anti-DDoS solutions and CDNs for added protection
11. Device Management:
o Use centralized systems for monitoring, updating, and configuring IoT devices
o Regularly scan and address vulnerabilities
12. Vendor and Policy Compliance:
o Work with vendors adhering to security standards
o Maintain a vulnerability disclosure policy with periodic assessments
13. Access Control and Isolation:
o Limit privileged access to hardware and firmware
o Run applications in containers or sandboxes for isolation
14. Deception and Monitoring:
o Deploy honeypots to detect malicious activity
o Use Runtime Application Self-Protection (RASP) to monitor IoT applications
15. Zero-Trust Model:
o Assume no device or user is trustworthy by default; continuously verify identity and
permissions
16. Blockchain and Emerging Technologies:
o Consider blockchain for secure, tamper-proof IoT data management
How to Prevent SDR-Based Attacks
Protecting IoT Devices from SDR-Based Attacks
IoT devices face threats from determined attackers equipped with specialized tools. To stay
ahead, proactive measures are essential to safeguard these devices before they are compromised.
Below are key methods to mitigate SDR-based threats:
1. Secure Signal Transmission
o Encrypt signals with robust and standardized encryption techniques to safeguard
communication from unauthorized access
2. Prevent Replay Attacks with Rolling Commands
o Avoid repeated use of identical commands by implementing a rolling window scheme.
This ensures each command is unique, reducing the risk of replay and brute-force attacks
3. Incorporate Synchronization and Preamble Nibbles
o Enhance protocol security by segregating command sequences with preamble and
synchronization nibbles. This prevents brute-force attacks that exploit overlaps in
command sequences, such as using a de Bruijn sequence to reduce necessary bits
4. Implement Anti-Jamming Measures
o Deploy anti-jamming technologies to detect and counteract unauthorized or disruptive
signal interference
5. Utilize Frequency Hopping
o Adopt Frequency-Hopping Spread Spectrum (FHSS) techniques to frequently change
radio frequencies, making it challenging for attackers to intercept or disrupt signals.
6. Strengthen Key Management
o Employ secure key management systems, such as Hardware Security Modules (HSMs),
to protect cryptographic keys and ensure secure operations in SDR-based
communications
7. Secure OTA Updates
o Ensure firmware updates and configuration changes are delivered over secure,
cryptographically signed channels to guarantee authenticity and integrity
General Guidelines for IoT Device Manufacturers
IoT device manufacturers must take necessary measures to implement the following
fundamental security practices:
● Utilize SSL/TLS protocols for secure communication
● Conduct mutual verification of SSL certificates and maintain a certificate revocation list
● Enforce the use of strong, secure passwords.
● Store credentials in secure, trusted storage instead of hardcoding them
● Ensure a simple and secure update process with a chain of trust
● Implement account lockout mechanisms to prevent brute force attacks after repeated failed
login attempts
● Harden devices against attacks by restricting unnecessary features and configurations
● Regularly remove unused tools and use whitelisting to permit only trusted applications
● Employ secure boot chains to validate all executed software
● Test new product features for potential security vulnerabilities before release
● Avoid risky functions like gets() using secure alternatives to prevent buffer overflow
vulnerabilities
● Integrate security measures into the IoT software development lifecycle
● Protect users’ data with clear data-sharing and transfer policies
● Provide consumers with guidelines for secure device configuration
● Equip devices with external hardware tamper alerts to ensure physical security
● Incorporate network security tools such as firewalls, intrusion detection systems, and
network segmentation
● Maintain transparency about security features and risks while offering clear channels for
reporting vulnerabilities
● Use secure, industry-standard communication protocols like MQTT, CoAP, or HTTPS for
data transmission
● Add hardware-based security components, such as Trusted Platform Modules (TPM) or
secure elements, for cryptographic key protection, secure boot, and tamper-resistant data
storage
OWASP Top 10 IoT Vulnerabilities Solutions
IoT Framework Security Considerations
Addressing security concerns during the design phase is essential to ensure IoT devices are
secure and well-protected. A critical aspect is creating a robust IoT framework that supports
secure device development. Ideally, the framework should include built-in security features by
default, eliminating the need for developers to add them later. The evaluation criteria for IoT
frameworks are divided into four sections, each addressing specific security aspects. Below are
the security considerations for IoT devices:

Edge
The edge represents the primary physical device in the IoT ecosystem. It interacts with the
environment and incorporates sensors, actuators, operating systems, hardware, networks, and
communication modules. Since edge devices are diverse and can be deployed under various
conditions, an effective framework should offer cross-platform compatibility, enabling
deployment and operation in any environment. Additional security measures for edge devices
include encrypted communication and storage, avoiding default credentials, enforcing strong
passwords, using updated components, and ensuring overall system integrity.

Gateway
The gateway is the entry point for edge devices to connect to the internet, linking smart devices
with cloud components. Often described as a communication aggregator, it facilitates secure
communication within a trusted local network and establishes secure connections with
untrusted public networks. Additionally, the gateway enhances the security of all connected
devices. As the central aggregation point for edge devices, it plays a vital security role in the IoT
ecosystem.
A well-designed gateway framework should include robust encryption protocols to ensure
secure communication between endpoints. Authentication mechanisms for edge components
must be equally robust, maintaining the same high-security standard across the framework.
Wherever feasible, the gateway should support multi-directional authentication, enabling
trusted communication between edge devices and the cloud. Furthermore, the gateway should
have automated update capabilities to address vulnerabilities proactively.
Cloud Platform
The cloud is the central hub for data aggregation and management in an IoT ecosystem. Access
to the cloud should be tightly controlled. The cloud component is more vulnerable to risks as
the focal point for most of the ecosystem's data. It also houses the Command-and-Control (C2)
system, a central computer that manages commands for distributing extensions and updates.
A secure framework for the cloud should incorporate encrypted communication, robust
authentication methods, a secure web interface, encrypted storage, and automatic updates to
ensure ongoing protection.

Mobile
The mobile interface plays a crucial role in an IoT ecosystem, especially in data collection and
management. Users can remotely access and interact with edge devices in their homes or
workplaces through mobile interfaces. Some mobile apps offer limited access to specific data
from edge devices, while others provide full control over the edge components. Given their
susceptibility to various cyber threats, the security of mobile interfaces requires careful
attention.
An effective framework for mobile interfaces should include strong user authentication, an
account lockout feature after several failed attempts, secure local storage, encrypted
communication channels, and protection of data transmitted across those channels.
IoT Hardware Security Best Practices
Securing IoT hardware is crucial in preventing many of the persistent attacks that occur at the
foundational level. However, variations in the manufacturing, development, and deployment
processes of IoT hardware across different manufacturers can introduce vulnerabilities.
Organizations can adopt the following strategies to protect their IoT hardware from common
attacks:
 Restrict Access Points
Minimize the potential for attack by avoiding adding unnecessary entry points, such as USB
ports, to IoT hardware. This reduces the risk of attackers exploiting open or unused ports.
Ensure that the least-accessed or open points are securely locked to prevent unauthorized
access to the device.
 Implement Tamper Detection
Introduce mechanisms to detect tampering with the hardware, such as physical damage or
unauthorized access at the board level. This can help identify attempts to steal, alter, or gain
access to components like the device's lid, chips, or internal parts. Adding a GPS tracker can
also assist in locating misplaced devices.
 Monitor Booting Process
Track the boot source to prevent attackers from interfering with the hardware startup
process, which could lead to boot-level attacks and provide unauthorized access. For higher-
value assets, consider integrating secure data storage solutions like Trusted Platform
Module (TPM) chips to safeguard data integrity.
 Apply Security Updates
Ensure timely and secure firmware updates, as IoT hardware is particularly vulnerable to
threats during pre and post patch update phases.
 Establish Effective Interface Management
During development, secure interfaces should be integrated into the IoT hardware to
prevent unauthorized access to APIs and libraries. APIs can pose a security risk by exposing
sensitive data about the device’s activities and functions at the user’s location.
 Prevent Unrestricted Access to Hardware
Provide robust physical protection for IoT hardware, especially since many devices are
deployed in public spaces or harsh environments. Secure all open ports or entry points with
dummy plugs to prevent unauthorized access or physical damage.
 Protect Authentication Keys
Ensure the secure storage of authentication keys, typically assigned by the cloud service,
used to verify each device's identity. Implement a strong security protocol to protect these
keys, as a breach could give attackers control over the hardware.
 Implement an Effective Event Logging System
Conduct regular security audits to establish a continuous event logging and monitoring
system for better incident management and response. Maintain detailed logs of all security
breaches to prevent future attacks on the hardware.
 Establish Robust Anti-Malware Protection
Deploy reliable third-party antivirus or anti-malware software to detect and prevent security
vulnerabilities at the hardware’s entry points.
 Enable Default Entry-Level Log Monitoring
Review entry logs regularly by activating the continuous logging feature offered by most
operating systems. This log monitoring helps prevent significant security issues at the entry-
level.
 Safeguard Device Access Credentials
Protect device access credentials by implementing various security measures, such as magic
numbers, encryption, and two-factor authentication.
 Isolate Devices from Power Supply Risks
Shield devices from electrical threats and surge attacks, such as rowhammer attacks, which
cause sudden voltage spikes that can damage RAM chips and other critical components.
 Implement a Root-of-Trust System
Establish a secure entry point for root-level access by utilizing a root-of-trust mechanism
that only restricts access to authorized and trusted users.
 Secure Legacy Devices with Modern Gateway Security
Introduce modern gateways into IoT networks containing legacy devices to apply advanced
security features without modifying or upgrading the devices.
 Harden Wireless Communication
Strengthen wireless communication interfaces (Wi-Fi, Bluetooth, and Zigbee) by applying
encryption, authentication, and access control measures to enhance security.
 Secure Debugging Interfaces
Deactivate or protect debugging interfaces (like JTAG and UART) to block unauthorized
access to the device’s internal components and sensitive data.
 Establish a Hardware-based Root of Trust
Create a hardware-based root of trust using Trusted Execution Environments (TEEs) or
secure enclaves to safeguard critical security processes and cryptographic functions from
tampering or compromise.
 Secure Handling of Sensor Data
Adopt secure practices for managing sensor data gathered by IoT devices, including data
validation, integrity checks, and privacy protection techniques.
 Integrate Hardware-based Intrusion Detection
Implement hardware-based intrusion detection systems, such as sensors or monitoring
circuits, to detect physical tampering, side-channel attacks, or unauthorized access to device
internals.
Countermeasures
Here are the countermeasures for securing communication data and TPMs in IoT hardware
units:
 Use Third-Party Authentication Software
Incorporate third-party authentication tools, such as Bitlocker drive encryption, to authenticate
data imported from external storage locations beyond the TPM perimeter.
 Apply Software Tools for TPM Communication
Software solutions like Nuvoton for IoT devices manage communication via interfaces like I2C
and SPI in TPM-equipped hardware.
 Bind Data Using TPM Keys
Secure data by binding it with a TPM-specific encryption key based on the RSA encryption
standard before transferring it.
 Implement Sealing and Unsealing during Updates
Apply the sealing and unsealing concept of hardware authentication during critical
computational updates, such as firmware updates or security patch installations.
 Use HMAC-based Secure Communication
Implement HMAC-key-based secure communication between the TPM-based IoT device and
the end user to protect data during transmission.
 Use Symmetric-Key Encryption for Low-Data Applications
Employ symmetric-key encryption for applications involving minimal data transmission outside
the TPM perimeter to maintain data authenticity and integrity.
 Verify Sender Authentication Before Decryption
Ensure sender authenticity before decrypting data by utilizing HMAC verification with TPM
devices and applying block-mode encryption algorithms like CBC or CFB.
 Use RSA-based Encryption for Data Integrity
Guarantee data integrity by applying RSA-based encryption and digital signatures to verify the
data.
 Store Keys in Non-Volatile Memory
Leverage TPMs to store encryption keys in Non-Volatile Random-Access Memory (NVRAM),
ensuring read/write capabilities during incidents like data loss caused by environmental stress
or attacks.
 Use Canonical Data Transfer Mode
Optimize data transfer using the canonical mode, which removes unnecessary bytes from
extended data-stream communications in TPM-enabled IoT devices.
 Implement Root-of-Trust Models
Use root-of-trust models like RT for Measurement (RTM) and RT for Verification (RTV) from
TPM devices to ensure secure booting and data transmission in IoT units.
 Ensure Perfect Forward Secrecy
Enable perfect forward secrecy by ensuring each communication session uses a unique session
key independent of long-term keys.
 Use Certificate-Based Authentication
Authenticate IoT devices and backend servers using certificate-based mechanisms before
establishing secure connections.
 Apply Remote Attestation via TPM
Utilize remote attestation features from TPMs to verify the integrity of IoT devices and evaluate
their security posture against remote servers or services.
 Use Authenticated Encryption with Associated Data (AEAD)
Implement AEAD algorithms, such as AES-GCM or AES-CCM, to provide confidentiality and
integrity for communication data.
 Leverage Hardware-Based Random Number Generators
Use hardware-based Random Number Generators (RNGs) to generate cryptographic keys and
Initialization Vectors (IVs) for encryption processes.
 Use Cryptographic Hardware Acceleration
Offload encryption and decryption operations from the CPU using cryptographic hardware
acceleration features in modern processors or dedicated cryptographic coprocessors.
 Implement Key Rotation Policies
Establish key rotation policies to periodically change the encryption keys used for
communication between IoT devices and backend servers.
Secure Development Practices for IoT Applications
Here are some secure development practices to safeguard IoT applications:
 Ensure Secure Boot
Verify that devices only run authenticated and validated code during the boot process to prevent
unauthorized firmware updates or modifications.
 Secure API Endpoints
Protect APIs with proper authentication and ensure thorough data validation to avoid
vulnerabilities like SQL injection, Cross-Site Scripting (XSS), and Cross-Site Request Forgery
(CSRF).
 Implement Threat Modeling
Assess potential security risks and threats specific to the IoT application and its ecosystem,
considering factors like data privacy, device authentication, and communication protocols.
 Follow Secure Coding Practices
Adhere to secure coding guidelines to avoid common issues such as buffer overflows, injection
attacks, and XSS vulnerabilities within the IoT application code.
 Perform Security Testing
Throughout the development process, carry out extensive security testing, penetration testing,
vulnerability scanning, and code reviews to identify and address security weaknesses.
 Secure Firmware or Software Updates
Implement secure Over-The-Air (OTA) mechanisms to safely deliver patches and firmware
updates to IoT devices, preventing unauthorized alterations and reducing the risk of
exploitation.
 Ensure Device Identity Management
Use unique identifiers and digital certificates for IoT devices to enable secure authentication
and establish trust within the IoT ecosystem.
 Implement Hardware Security
Leverage hardware-based security features such as Trusted Platform Modules (TPM), secure
elements, or Hardware Security Modules (HSM) to store cryptographic keys, perform secure
operations, and protect sensitive data.
 Allow Code Signing
Digitally sign firmware, software updates, and application codes to verify their authenticity and
integrity before installation, reducing the risk of unauthorized changes or malware insertion.
 Implement Runtime Protection
Adopt runtime protection strategies like monitoring code execution, preventing stack
overflows, and ensuring memory safety to detect and block software vulnerabilities at runtime.
 Ensure Secure Cloud Integration
Implement appropriate authentication, access control, and data encryption measures to protect
sensitive data stored or processed in the cloud and ensure secure connectivity with cloud
services.
 Utilize Secure Communication Protocols
Use secure communication protocols, such as MQTT with TLS/SSL, for device-to-cloud
communication, ensuring data confidentiality, integrity, and secure exchange between IoT
devices and backend servers.
IoT Device Management
IoT device management enables security professionals to monitor and control IoT devices
remotely. Platforms like Azure IoT Central, Oracle Fusion Cloud Internet of Things (IoT), and
Predix facilitate these tasks, allowing professionals to perform actions such as firmware updates
from a distance. Additionally, IoT device management strengthens security by granting
permissions and implementing measures to protect against vulnerabilities. This approach plays
a crucial role in mitigating IoT-based threats through:
● Authentication: Ensuring only trusted devices with valid credentials are connected
● Configuration: Maintaining device functionality and performance, including resetting
factory settings during decommissioning
● Monitoring: Identifying flaws, diagnosing operational issues, and addressing software bugs
through program logs
● Maintenance: Applying security updates and patches to remote devices regularly
These practices collectively enhance device security and reduce the risk of IoT-related attacks.
IoT Device Management Solutions
Security experts, IT administrators, and IT managers utilize IoT device management tools to
streamline the onboarding, organization, monitoring, and control of IoT devices. Below are
some key IoT device management platforms:
● Azure IoT Central
Azure IoT Central is a scalable Software-as-a-Service (SaaS) platform designed to simplify the
development and management of IoT solutions. It provides an intuitive way to connect, oversee,
and manage large-scale IoT devices. This platform reduces the complexities of initial IoT
deployments and helps lower operational costs, management efforts, and project overheads.

Figure 18-68: Azure IoT Central

Additional IoT device management solutions include:


● Oracle Fusion Cloud Internet of Things (IoT) (https://www.oracle.com)
● Golioth (https://golioth.io)
● AWS IoT Device Management (https://aws.amazon.com)
● IBM Watson IoT Platform (https://www.ibm.com)
● openBalena (https://www.balena.io)
IoT Security Tools
The Internet of Things (IoT) is a network of connected devices and a rapidly advancing and
intricate technology. To address and mitigate potential risks, robust security solutions must be
implemented to safeguard IoT devices. IoT security tools are crucial in minimizing
vulnerabilities and protecting devices and networks from cyber threats.
SeaCat.io
SeaCat.io is a security-focused SaaS platform designed to support IoT products with reliability,
scalability, and robust protection. It offers tools for centralized management of connected
devices, remote access, and real-time monitoring. Security experts utilize SeaCat.io to automate
bug-fix updates, secure users with authorized cryptographic measures, ensure compliance with
regulations, and maintain malware-free devices. Additionally, it helps prevent unauthorized
control of IoT devices, protecting them from being exploited as part of botnets.
These features make SeaCat.io an invaluable solution for ensuring IoT security at scale.

Figure 18-69: SeaCat.io

Armis Centrix™
Armis Centrix™ is a comprehensive tool that assists security professionals in managing,
protecting, and optimizing IoT assets, systems, and processes within their environment. It
addresses IoT vulnerabilities by implementing tailored security measures aligned with industry
standards and regulatory requirements. The platform identifies outdated operating systems and
ensures timely updates to maintain device security. Armis Centrix™ proactively detects
signature-based threats and Indicators of Compromise (IOCs), offering early warning against
potential attacks. It also collects and analyzes forensic threat data at every stage of an incident,
enabling security teams to make informed, data-driven decisions for prioritizing responses
effectively.

Figure 18-70: Armis CentrixTM

Additional IoT security tools and solutions include:


● FortiNAC (https://www.fortinet.com)
● Microsoft Defender for IoT (https://www.microsoft.com)
● Symantec Critical System Protection (https://www.broadcom.com)
● Cisco Industrial Threat Defense (https://www.cisco.com)
● AWS IoT Device Defender (https://aws.amazon.com)
● Forescout (https://www.forescout.com)
● NSFOCUS Anti-DDoS System (https://nsfocusglobal.com)
● Azure Sphere (https://www.microsoft.com)
● Overwatch (https://overwatchsec.com)
● Barbara (https://www.barbara.tech)
● Sternum (https://sternumiot.com)
● Asimily (https://asimily.com)
● ByteSweep (https://gitlab.com)
● Entrust IoT Security (https://www.entrust.com)
● IOT ASSET DISCOVERY (https://securolytics.io)

OT Hacking
OT Concepts and Attacks
Operational Technology (OT) holds significant importance in today's interconnected world, as
it comprises systems of devices functioning together as unified or cohesive units. For instance,
in telecommunications, OT facilitates data transfer within the electrical grid and supports
financial transactions between electricity providers and consumers. OT combines hardware and
software to oversee, operate, and manage industrial process assets. To effectively understand
OT hacking, it is crucial first to grasp its fundamental principles. This section explores key
concepts related to OT.
As security threats evolve and organizations increasingly rely on OT, it becomes vital to
prioritize OT security and implement effective measures to address challenges arising from the
integration of OT and IT systems. The discussion also highlights various OT-specific threats and
attack methods, such as compromising industrial networks, targeting Human-Machine
Interfaces (HMIs), exploiting side channels, tampering with Programmable Logic Controllers
(PLCs), and manipulating industrial equipment through RF remote controllers.
What is OT?
Operational Technology (OT) integrates hardware and software to monitor and control
industrial processes, enabling changes in physical systems. It encompasses devices like pumps,
switches, lights, sensors, surveillance cameras, elevators, robots, valves, and climate control
systems. Any platform that processes or evaluates operational data—spanning components like
electronics, telecommunications, and computer systems—can fall under the umbrella of OT.
This technology plays a crucial role across various industries, including manufacturing, mining,
healthcare, transportation, oil and gas, defense, utilities, and more, ensuring the safe operation
of physical systems within networks.
OT includes Industrial Control Systems (ICS), such as Supervisory Control and Data Acquisition
(SCADA), Programmable Logic Controllers (PLC), Remote Terminal Units (RTU), Distributed
Control Systems (DCS), and other specialized network systems designed for industrial
management. Unlike IT, OT hardware and protocols are built with distinct approaches, often
relying on older technologies. This reliance on outdated software and equipment increases their
susceptibility to cyber threats, as creating updates or patches for these systems is particularly
challenging.

Figure 18-71: Devices Connected to an OT Network


Figure 18-72: Components of OT

Essential Terminology
Below is an overview of key terms commonly associated with OT systems:
● Assets: The components of OT systems are collectively called assets. These include physical
assets such as sensors, actuators, servers, workstations, network equipment, and PLCs. OT
systems also consist of logical assets, encompassing physical components' representations
and functionalities, such as process flow diagrams, program logic, databases, firmware, or
firewall configurations.
● Zones and Conduits: Zones and conduits refer to a network segmentation method to
isolate systems and assets. This approach enhances access control by dividing networks into
distinct zones connected through secured conduits.
● Industrial Network and Business Network: OT systems are generally composed of
automated control systems linked together to achieve specific business goals. These
interconnected systems form an industrial network. On the other hand, a business network
provides the information infrastructure required for business operations. Establishing
communication between industrial and business networks is often essential for seamless
operations.
● Industrial Protocols: OT systems utilize specialized communication protocols. These
include proprietary ones like S7, CDA, and SRTP or non-proprietary ones like Modbus, OPC,
DNP3, and CIP. These protocols facilitate serial and support communication over Ethernet
using IP alongside transport protocols like TCP and UDP. These protocols are classified as
applications since they function at the application layer.
● Network Perimeter/Electronic Security Perimeter: The network perimeter defines the
outer boundary of a network zone, representing a secure grouping of assets. This perimeter
is a barrier between a zone's internal and external environments, where cybersecurity
measures are typically applied. An electronic security perimeter defines the boundary
separating secure zones from unsecured areas.
● Critical Infrastructure: This term refers to essential physical or logical systems and assets
whose compromise, failure, or destruction would have a profound impact on public safety,
security, economic stability, or health services.
Introduction to ICS
Industrial Control Systems (ICS) are vital to industrial operations and critical infrastructure
across various sectors. These systems are information platforms that manage and support
industrial activities, including production, manufacturing, product handling, and distribution.
ICS encompasses various control systems and related equipment, such as devices, networks,
and controls, designed to automate and operate numerous industrial processes.
An ICS integrates various control systems, including Supervisory Control and Data Acquisition
(SCADA), Distributed Control Systems (DCS), Basic Process Control Systems (BPCS), Safety
Instrumented Systems (SIS), Human-Machine Interfaces (HMI), Programmable Logic
Controllers (PLC), Remote Terminal Units (RTU), and Intelligent Electronic Devices (IED). Key
components of ICS include sensors, controllers, and actuators—ranging from mechanical to
hydraulic, electrical, and pneumatic—that work together to accomplish industrial objectives.
The process generates the desired output within an ICS, while the control segment provides the
instructions needed to achieve this output. Control can be fully automated or involve human
participation in the process loop. ICS operations can function in three modes: open-loop,
closed-loop, or manual-loop mode.
● Open-Loop: In this configuration, the system's output is determined by predefined
settings, operating independently of feedback
● Closed-Loop: The output influences the input, allowing the system to adjust and achieve
the desired goal
● Manual-Loop: This mode places full control of the system in the hands of human operators
The controller within an ICS is tasked with ensuring operations align with specified
requirements. ICS systems typically consist of multiple control loops, Human-Machine
Interfaces (HMIs), and tools for remote diagnostics and maintenance. These remote
management tools leverage various networking protocols to function effectively.
ICS systems find extensive application across numerous industries, including electricity
generation and distribution, water treatment and supply, oil and gas pipelines, chemical and
pharmaceutical manufacturing, paper production, and food and beverage processing. In certain
sectors, ICS systems are physically distributed across multiple sites with interdependent
processes. To facilitate seamless coordination among these distributed systems, communication
protocols play a crucial role in enabling efficient interaction.
Figure 18-73: Components of an ICS

Figure 18-74: ICS Architecture

Components of an ICS
An Industrial Control System (ICS) encompasses many networks and systems essential for
managing and overseeing industrial processes. The specific operation of each type of ICS varies
depending on the complexity and functionality of the control actions. ICSs can be categorized
into several types, with the most commonly used being:
Distributed Control System (DCS)
A DCS is designed to manage production systems within a single geographic location. It is
primarily utilized for large, intricate, and distributed processes found in industries such as
chemical manufacturing, nuclear plants, oil refineries, water and sewage treatment, power
generation, and pharmaceutical and automobile production. A DCS is a highly sophisticated,
large-scale control system, often tailored to perform specific tasks within an industry. It features
a centralized supervisory control unit that oversees multiple local controllers, thousands of
Input/Output (I/O) points, and various field devices involved in the production process.
A DCS uses various feedback and feedforward loops to achieve process control, with key product
conditions set according to target parameters. It operates with a centralized supervisory loop,
such as SCADA or MTU, that connects localized controllers like RTU/PLC to execute the
necessary tasks for the overall process. Redundancy is incorporated at every level, from I/O
controllers to the network, ensuring continuous operations during a processor failure.
The main advantage of a DCS in industrial settings is its adaptability and flexibility, which allow
for efficient control of distributed field devices and their respective stations. Additionally, a DCS
is scalable, enabling it to be initially installed as either a large integrated or modular system that
can expand according to requirements. DCS technology continues to evolve, integrating new
advancements such as wireless systems, remote transmission, data logging and historians, and
embedded web servers.

Figure 18-75: DCS Architecture

Supervisory Control and Data Acquisition (SCADA)


SCADA is a centralized control system that oversees and manages industrial operations and
infrastructure. Many organizations utilize SCADA to automate complex industrial processes,
track real-time trends, and identify and resolve issues. Typically, SCADA systems are distributed
over vast geographical areas, making them essential in oil and gas transportation, wastewater
management, pipeline operations, telecommunications, power grids, building automation, and
public transportation systems.
As a centralized system, SCADA facilitates supervisory control while enabling real-time data
acquisition from various assets involved in industrial processes. The system comprises hardware
and software components that collect and transmit data to monitor and control processes at
local and remote sites. The collected data is stored in long-term storage systems, such as data
historians, which assist operators in interpreting the data and adjusting system parameters.
These adjustments, or setpoints, allow the system to respond efficiently to abnormal conditions
by issuing commands or sending alerts to operators.
SCADA systems integrate data acquisition, transmission, and Human-Machine Interface (HMI)
software to offer centralized control and monitoring of multiple process inputs and outputs.
The system gathers information from field devices and relays it to a central computer. This data
is then presented to operators in graphical or textual formats, enabling them to monitor and
control the entire system from a central location in real-time.
The SCADA system architecture includes hardware components such as a control server
(SCADA-MTU) and communication devices like network cables, radio systems, telephone lines,
and cables. It also features a network of geographically dispersed field sites containing devices
like PLCs and RTUs to monitor and manage industrial equipment. The information from the
RTU is processed and managed by the control server, while the RTU or PLC is responsible for
overseeing and controlling the field devices. SCADA software is designed to specify which
parameters need monitoring, the timing for monitoring, and the acceptable parameter limits.
It also outlines the necessary actions if parameters exceed the predefined ranges.
An Intelligent Electronic Device (IED) may either directly collect data and transmit it to the
control server or be directed by a local RTU to collect and send it. The IED contains a
communication interface to monitor and control various sensors and devices. IEDs may be
controlled directly by the control server or feature local programming that allows them to
operate autonomously without requiring server intervention. While SCADA systems are built
with redundancy for fault tolerance, this redundancy may not fully protect the system from
malicious attacks.

Figure 18-76: SCADA Architecture

Programmable Logic Controller (PLC)


A Programmable Logic Controller (PLC) is a real-time digital computer designed for industrial
automation. Unlike standard digital computers, PLCs are valued in various industrial control
systems for their robust design, ease of programming, sequential control, user-friendly
hardware, timers and counters, and dependable control functions. Built to withstand harsh
industrial environments, PLCs are commonly used in industries like steel, automotive, energy,
chemicals, glass, paper, and cement manufacturing. A small solid-state computer, PLCs can be
programmed to perform specific tasks. The instructions stored in a PLC allow it to carry out
functions such as logic operations, timing, counting, I/O control, communication, arithmetic,
and file or data processing. PLCs have largely replaced older systems like drum sequencers,
hard-wired relays, and timers. They continuously monitor sensor input values and generate the
necessary outputs for controlling actuators.
A PLC system is made up of three key modules:
1. CPU Module: This module includes the central processor and memory. The processor
handles data calculations and processes by receiving inputs and generating the
corresponding outputs. The memory consists of RAM and ROM, with RAM storing user
programs and ROM containing the operating system, drivers, and application programs.
PLCs also feature retentive memory, preserving user programs and data during power
outages. This allows the system to resume operation once power is restored, eliminating the
need to reprogram the processor.
2. Power Supply Module: This module provides the necessary power for the CPU and I/O
modules by converting AC power into DC. It is essential for the operation of the PLC system.
The module outputs 5V DC to power the PLC's internal circuitry. In some cases, it provides
24V DC to power sensors and actuators.
3. I/O Modules: The input and output modules connect sensors and actuators to the system,
enabling the monitoring and controlling of real-time parameters such as pressure,
temperature, and flow. There are different types of I/O modules:
o Digital I/O Module: Used to connect digital sensors and actuators that
operate with ON/OFF switching. These modules handle multiple digital
inputs and outputs and support AC and DC voltages.
o Analog I/O Module: Used to connect sensors and actuators that generate
analog signals. This module includes an analog-to-digital converter to convert
analog signals into digital form, which the CPU processes.
o Communication I/O Module: Facilitates the exchange of information
between the communication network and the CPU, particularly when the CPU
is located remotely.
A PLC's primary function is to automate machinery and processes without human intervention,
making it vital in industries such as manufacturing, production, and growth.

Figure 18-77: PLC Architecture

Basic Process Control System (BPCS)


A Basic Process Control System (BPCS) controls and monitors industrial processes. It responds
to input signals from processes and related equipment to generate output signals that allow the
processes and equipment to function according to a predetermined control strategy. BPCS
systems are dynamic and highly flexible, enabling them to adjust to changing process
conditions. They are used in various control loops, such as temperature, batch, pressure, flow,
feedback, and feedforward loops, in industries like chemicals, oil and gas, and food and
beverages. BPCSs are essential in industrial settings, serving as the first defense against unsafe
or hazardous conditions that could damage equipment. These systems are also designed to push
performance limits to achieve desired outcomes. BPCSs do not typically include diagnostic
functions to detect system faults, unlike safety control systems. Still, when properly designed,
they effectively address a wide range of operational and monitoring challenges in industrial
environments.
Here are some key functions provided by a Basic Process Control System (BPCS):
● Enables trending and logs alarms/events
● Offers an operator interface (HMI) for monitoring and controlling the system
● Manages processes to optimize plant operations and improve product quality
● Generates production data reports
● Coordinates batch process steps' sequencing, timing, and synchronization to maintain
consistent quality and efficiency
● It includes essential safety interlocks to prevent equipment operation under unsafe
conditions, ensuring the process's and personnel's protection
● Manages the storage and retrieval of recipes, such as formulas, process steps, and production
parameters, making it easier to replicate batch processes
● Integrates with other business and engineering systems, bridging the gap between plant
floor activities and business management

Figure 18-78: BPCS Architecture

Safety Instrumented System (SIS)


A Safety Instrumented System (SIS) is an automated control system that protects the
manufacturing environment during a hazardous incident. These systems monitor and execute
specific control functions to shut down or bring the monitored system to a predefined safe state,
minimizing the negative impact of such incidents. SIS plays a vital role in a risk management
strategy, applying multiple layers of protection to prevent critical processes from reaching
unsafe conditions. Examples of SIS include fire and gas detection systems, safety interlock
systems, and safety shutdown systems.
An SIS takes control in industrial settings when a Basic Process Control System (BPCS) fails to
maintain normal operational parameters. If the BPCS operates beyond safe limits, the SIS
activates to detect and respond to the situation, ensuring the process is either preserved or
transitioned to a safe state, such as shutting down equipment or the process itself. The final
layer of protection includes devices like relief valves, rupture disks, and flare systems, which act
before the process reaches unsafe limits. The events triggered and the actions taken by the SIS
system are depicted in the diagram:

Figure 18-79: Layers of Protection Provided by SIS Systems

The functional requirements and efficiency of the work performed by an SIS can be determined
through methods like Hazard and Operability Studies (HAZOP), Layers of Protection Analysis
(LOPA), and risk graphs. The SIS operates independently from other control systems and is
composed of sensors, logic solvers, and final control elements that ensure the process remains
safe by carrying out the following tasks:
● Field sensors: These devices gather data on process parameters such as temperature,
pressure, and flow to assess whether the equipment functions within safe limits. Sensors can
include pneumatic, electric switches, or smart transmitters.
● Logic solvers: These systems evaluate the data from the sensors and decide on the
appropriate action. They ensure safety in failsafe and fault-tolerant scenarios by acting as
controllers that process the sensor signals and provide outputs to the final control elements.
● Final control elements: These components carry out the actions determined by the logic
solvers to bring the system to a safe state. Typically, these include pneumatically controlled
on/off valves managed by solenoid valves.
Since no component is entirely immune to failure, it is crucial for industries to continuously
test the SIS and assess its cybersecurity environment to ensure proper functioning. The primary
goal of this assessment is to maintain the SIS's safety and integrity and ensure it operates at its
design standards.
Figure 18-80: SIS Architecture

IT/OT Convergence (IIOT)


IT/OT convergence refers to integrating Information Technology (IT) systems with Operational
Technology (OT) monitoring systems. Closing the gap between IT and OT can enhance overall
business performance, leading to faster and more efficient outcomes. This convergence involves
not only merging technologies but also aligning teams and operations. Traditionally, IT and OT
teams work separately within their respective domains. For example, IT teams focus on
programming, system updates, and protecting networks from cyber threats.
In contrast, OT teams are responsible for maintaining and managing equipment and personnel
in industrial settings. For effective IT/OT convergence, both teams need to understand each
other's roles and operations. This does not mean swapping IT engineers for field engineers but
rather creating a collaborative bridge between the two to improve security, efficiency, quality,
and productivity.
Benefits of merging OT with IT
IT/OT convergence paves the way for smart manufacturing, also known as Industry 4.0, where
IoT applications play a key role in industrial operations. IoT in monitoring areas like supply
chains, manufacturing, and management systems is called the Industrial Internet of Things
(IIoT). The integration of IT and OT offers several advantages, including:
● Improved Decision Making: Integrating OT data with business intelligence systems
enhances decision-making capabilities.
● Boosted Automation: Merging IT and OT optimizes business processes and industrial
control, driving better automation.
● Faster Business Output: IT/OT convergence streamlines development processes, leading
to faster business output.
● Cost Reduction: It helps lower both technological and organizational costs.
● Risk Mitigation: The integration improves productivity, security, and reliability, ensuring
scalability in the process.
● Increased Agility: Unified systems make organizations more adaptable to changes in
market conditions or operational demands, offering a competitive edge in dynamic
industries.
● Predictive Maintenance: IT systems can analyze data from OT devices to predict
equipment failures, reducing downtime and maintenance costs and extending equipment
lifespan.
● Enhanced Quality Control: The integration enables more thorough monitoring and
control of quality standards across production lines, improving product quality and
consistency.
● Simplified Compliance and Reporting: IT/OT convergence streamlines compliance with
regulatory requirements by enhancing data collection, analysis, and reporting, ensuring
data accuracy, traceability, and audit accessibility.
● Scalability: Integrated IT and OT systems can be easily scaled to meet evolving business
needs, supporting growth and adaptation without requiring significant infrastructure
changes.

Figure 18-81: IT/OT Convergence

The Purdue Model


The Purdue model is based on the Purdue Enterprise Reference Architecture (PERA). This
widely adopted framework outlines the key components and their interconnections within ICS
networks. Also referred to as the Industrial Automation and Control System (IACS) reference
model, the Purdue model divides the system into three main zones: the manufacturing zone
(OT), the enterprise zone (IT), and a Demilitarized Zone (DMZ) that separates the OT and IT
systems. The DMZ limits direct communication between the two, helping to contain potential
security breaches within this layer and ensuring continuous production. These three zones are
further broken down into several operational levels, each of which is described below:
Figure 18-82: Purdue Model

Enterprise Zone (IT Systems)


The enterprise security zone belongs to the IT domain, where business operations such as
supply-chain management and scheduling are handled through systems like SAP and ERP. It
also houses data centers, user access, and cloud connections. This zone is divided into two
levels:
 Level 5 (Enterprise Network): This corporate-level network manages business operations
like Business-to-Business (B2B) and Business-to-Customer (B2C) services, internet
connectivity, and network management. It also aggregates data from individual plant
subsystems, providing reports on inventory and overall production status.
 Level 4 (Business Logistics Systems): This level supports the plant’s production process
through IT systems. It manages production schedules, planning, and other logistics,
including application servers, file servers, database servers, supervisory systems, and email
clients.
Manufacturing Zone (OT Systems)
This zone encompasses all devices, networks, control, and monitoring systems involved in
production. It consists of four levels:
 Level 3 (Operational Systems/Site Operations): This level manages production, plant
monitoring, and control functions, ensuring that production workflows and desired product
output are maintained. It includes plant performance management, scheduling, batch
management, quality assurance, data historians, MES/MOMS, laboratories, and process
optimization. Data from lower levels is collected here for transfer or action by higher levels.
 Level 2 (Control Systems/Area Supervisory Controls): At this level, physical processes
are supervised and controlled, utilizing control systems like DCSs, SCADA software, HMIs,
real-time software, and other supervisory control systems such as PLC line control and
engineering workstations.
 Level 1 (Basic Controls/Intelligent Devices): This level handles the analysis and
modification of physical processes. Basic control operations include starting motors,
opening valves, and moving actuators. Systems at this level include analyzers, process
sensors, IEDs, PLCs, RTUs, PID controllers, EUCs, and VFDs. PLCs serve as control
functions, whereas they function as supervisory tools at Level 2.
 Level 0 (Physical Process): This level defines the physical manufacturing process, where
products are made. Higher-level systems manage operations at this level, often referred to
as EUC. It includes devices, sensors (e.g., speed, temperature, pressure), actuators, and
industrial equipment critical for production. Any malfunction in these devices can disrupt
operations.
Industrial Demilitarized Zone (IDMZ)
The IDMZ acts as a security barrier between the manufacturing zone (OT) and the enterprise
zone (IT), providing a controlled and secure network connection between the two. This zone is
designed to inspect the overall system architecture, containing errors or intrusions that may
compromise operations while allowing production to continue. IDMZ systems include domain
controllers, database replication servers, and proxy servers.
OT Technologies and Protocols
Industrial network protocols enable real-time connectivity and data exchange between
industrial systems and zones. These protocols are implemented throughout the ICS network in
different industries. For a security engineer to comprehend an industrial network, it is essential
to understand the network infrastructure's protocols. The primary communication technologies
and protocols within the OT network, as outlined by ISA-95 in the Purdue model, include:

Figure 18-83: OT Technologies and Protocols Over the Purdue Model

Protocols Used in Levels 4 and 5


Protocols used in Levels 4 and 5 include:
 DCOM: Distributed Component Object Model (DCOM) is a Microsoft technology that
enables secure and reliable communication between software components over a network.
 FTP/SFTP: FTP connects to a specific server to upload or download files, while SFTP
establishes a secure connection by verifying the client's identity before exchanging
information.
 GE-SRTP: Service Request Transport Protocol (GE-SRTP), developed by GE Intelligent
Platforms, transfers data from PLCs, facilitating the conversion of digital commands into
physical actions on select GE PLCs.
 IPv4/IPv6: IPv4 is a connectionless protocol used in packet-switched networks, while IPv6
provides end-to-end transmission of datagrams across multiple IP networks.
 OPC UA: This protocol enables secure, reliable, and platform-independent data exchange
between manufacturing devices and enterprise systems.
 TCP/IP: TCP/IP is a suite of communication protocols that facilitate the interconnection of
networking devices over the internet.
 SMTP, HTTP/HTTPS: These standard internet protocols are used for email
communication, file transfers, and data transmission within enterprises.
 Wi-Fi: Wi-Fi is a wireless networking technology commonly used in Local Area Networks
(LAN). The 802.11n standard offers speeds up to 600 Mbps and a range of around 50 meters.
Protocols Used in Level 3
Protocols used in Level 3 include:
 CC-Link: Control and Communications Link (CC-Link) is an open industrial network that
facilitates communication between devices from various manufacturers, commonly used in
machine, process control, and building automation.
 HSCP: Hybrid Secure Copy Protocol (HSCP) is designed to transmit large files at high
speeds over long distances and wideband networks.
 ICCP (IEC 60870-6): Inter-Control Center Communications Protocol (ICCP) (IEC 60870-6)
defines standards for ICS or SCADA communication in power system automation.
 IEC 61850: IEC 61850 is a protocol enabling communication and interoperability between
Intelligent Electronic Devices (IEDs) at electrical substations.
 IEC 60870-5-104: This protocol is used in telecontrol systems for electrical engineering and
power system automation, particularly in utilities.
 ISA/IEC 62443: ISA/IEC 62443 provides a flexible framework to address and mitigate
current and potential security vulnerabilities in industrial automation and control systems.
 Modbus: Modbus is a serial communication protocol used with PLCs for device
communication over a shared network.
 NTP: Network Time Protocol (NTP) synchronizes clocks between computer systems over
packet-switched networks with variable latency.
 Profinet: Profinet is a communication protocol for exchanging data between controllers
like PLCs and devices like RFID readers.
 SuiteLink: SuiteLink is a TCP/IP-based protocol running as a service on Windows operating
systems. It is used primarily in industrial applications requiring high throughput, time
precision, and quality.
 Tase-2: Tase-2 (IEC 60870-6) is an open communication protocol that allows the exchange
of time-sensitive information between control systems across WAN and LAN networks.
 ControlNet: ControlNet is a real-time control protocol for collecting data and sending
instructions to field devices, offering high-speed transmission in noisy environments.
 Profibus PA/DP: Profibus Process Automation (PA) and Profibus Decentralized
Peripherals (DP) are used for plant-level automation. DP manages sensors and actuators via
a central controller, while PA monitors measuring equipment through process control
systems.
Protocols Used in Level 2
Protocols used in Level 2 include:
 6LoWPAN: IPv6 over Low Power Personal Area Networks (6LoWPAN) is an internet
protocol designed for communication between small, low-power devices with limited
processing capabilities, typically used in home and building automation systems.
 DNP3: Distributed Network Protocol 3 (DNP3) connects various components within process
automation systems.
 DNS/DNSSEC: Domain Name System Security Extensions (DNSSEC) provide methods to
authenticate DNS response data, ensuring the integrity and security of information received
from DNS.
 FTE: Fault-Tolerant Ethernet (FTE) ensures rapid network redundancy, with each node
connected to a single LAN through two network interfaces for added reliability.
 HART-IP: The HART-IP protocol facilitates the integration of WirelessHART gateways and
HART multiplexers, allowing efficient digital data transmission.
 IEC 60870-5-101/104: This protocol extends the IEC 101 standard with modifications in the
transport, network, link, and physical layers, enabling communication between control
stations and substations over standard TCP/IP networks.
 SOAP: Simple Object Access Protocol (SOAP) is a messaging protocol with a defined set of
rules for transferring data between a client and server using XML messages.
 DeviceNet: DeviceNet connects basic industrial devices like sensors and actuators to
higher-level devices, such as PLCs, running on Controller Area Network (CAN) technology.
 AS-Interface (AS-i): AS-i is a simple, cost-effective network that links binary devices like
actuators and sensors in automation systems.
Protocols Used in Level 0 and 1
Protocols used in Level 0 and 1 include:
 BACnet: Building Automation and Control Network (BACnet) is a communication protocol
for building automation systems, adhering to standards like ASHRAE, ANSI, and ISO 16484-
5.
 EtherCAT: Ethernet for Control Automation Technology (EtherCAT) is a fieldbus system
based on Ethernet, suitable for both hard and soft real-time computing needs in automation.
 CANopen: CANopen is a high-level communication protocol built on the Controller Area
Network (CAN) protocol, often used in embedded networks like vehicle communication
systems.
 Crimson: Crimson is a programming platform used across various Red Lion products,
including G3 and G3 Kadet HMIs, Data Station Plus, Modular Controller, and Productivity
Station.
 DeviceNet: DeviceNet is a version of the Common Industrial Protocol (CIP) used in
automation to connect control devices and facilitate data exchange.
 Zigbee: Zigbee is a short-range communication protocol based on the IEEE 203.15.4
standard. It is designed for low-data-rate, intermittent data transfer in confined areas within
10–100 m.
 ISA SP100: ISA SP100 is a committee that sets the industrial wireless standard ISA100 used
in the process automation and manufacturing industries.
 MELSEC-Q: MELSEC-Q offers an open, integrated network environment that connects
various automation networks, including CC-Link IE, high-speed, and large-capacity
Ethernet-based systems.
 Niagara Fox: Niagara Fox is a building automation protocol for communication between
Niagara software systems developed by Tridium.
 Omron Fins: Omron Fins is used by PLCs for data transfer and other services with remote
PLCs over Ethernet and can also be utilized by devices like FieldServer.
 PCWorx: PCWorx is used in various ICS components, enabling the integration of multiple
ICS protocols and common TCP/IP protocols.
 Profibus: Profibus is a more advanced protocol than Modbus, designed to resolve
interoperability issues. It is commonly used in process and factory automation.
 Sercos II: Sercos II is a serial real-time communication system with a digital drive interface
for industrial machines, especially in high-performance motion control applications.
 S7 Communication: S7 Communication is a Siemens proprietary protocol for data
exchange between Siemens S7-300/400 PLCs, programming, and accessing PLC data via
SCADA.
 WiMax: Worldwide Interoperability for Microwave Access (WiMax) is based on IEEE
802.16. It provides wireless metropolitan area networks with frequencies between 2.5 GHz
and 5.8 GHz and offers transfer rates of 40 Mbps.
 FOUNDATION Fieldbus: FOUNDATION Fieldbus is a digital communication protocol
used in control systems, particularly in process industries, to link field instruments like
sensors and actuators with control systems.
Challenges of OT
Operational Technology (OT) is critical in various essential infrastructure sectors, such as power
plants, water utilities, and healthcare. Unfortunately, many OT systems rely on outdated
software versions and obsolete hardware, making them susceptible to attacks like phishing,
espionage, ransomware, etc. These attacks can severely disrupt products and services. To
mitigate these vulnerabilities, OT systems must be carefully assessed for weaknesses using
appropriate security tools and strategies. The following challenges and risks contribute to the
vulnerabilities of OT systems:
 Lack of visibility: Gaining broad cybersecurity visibility within OT networks enhances
security and allows faster response to potential threats. However, many organizations lack
clear visibility, which makes it difficult for security teams to identify abnormal behaviors or
threat signatures.
 Plaintext passwords: Many industrial networks still use weak or plaintext passwords,
which leads to poor authentication practices and leaves systems exposed to cyber
reconnaissance attacks.
 Network complexity: OT networks are often complex, involving numerous devices with
varying security needs, which complicates their management and protection.
 Legacy technology: OT systems frequently rely on older technologies that lack essential
security measures like encryption and password protection, making them vulnerable to
various attacks. Additionally, applying modern security practices can be challenging.
 Absence of antivirus protection: Industries using outdated systems and legacy
technology often lack antivirus solutions that automatically update signatures, exposing
them to malware infections.
 Shortage of skilled security professionals: The cybersecurity talent gap presents a
significant risk, as not enough trained professionals are available to detect threats and
implement new security measures across networks.
 Rapid technological changes: Keeping up with technological advancements is a major
challenge in cybersecurity and slow digital transformation can expose OT systems.
 Outdated systems: Many OT devices, such as Programmable Logic Controllers (PLCs), run
on outdated firmware, making them highly vulnerable to modern cyberattacks.
 Inefficient modernization: As OT needs to grow, keeping systems updated with the latest
technologies becomes challenging. Upgrading and patching legacy components in OT
systems can take years, negatively impacting operations.
 Insecure connections: OT systems often communicate over public Wi-Fi and unencrypted
Wi-Fi in the IT network, making them prone to man-in-the-middle attacks.
 Presence of rogue devices: Many industrial networks have unauthorized or unknown
devices connected, which attackers can easily exploit.
 IT and OT convergence: OT systems are increasingly connected to corporate IT networks,
which exposes them to a wider range of malware and insider threats. Additionally, IT
security teams may not be well-versed in OT systems and protocols.
 Organizational challenges: Many organizations use different security architectures for IT
and OT, creating gaps in security management that attackers can exploit.
 Custom production networks and proprietary software: Industries often rely on unique
hardware and software configurations based on specific operational needs and industry
standards. The use of proprietary software complicates updates and firmware patching, as
multiple vendors control it.
 Vulnerable communication protocols: OT systems often use protocols like Modbus and
Profinet to control and connect various devices, such as sensors and controllers. These
protocols lack essential security features like authentication and anomaly detection, making
them vulnerable to cyberattacks.
 Remote management protocols: Industrial sites use remote protocols like RDP, VNC, and
SSH. Suppose an attacker gains access to the OT network. In that case, they can exploit
these protocols to manipulate the equipment's configuration and operation.
 Insufficient segmentation: Poor network segmentation in industrial environments allows
attackers to move laterally across networks once they gain initial access, increasing the
scope of potential damage.
 Physical security: OT systems are sometimes located in remote or unsecured areas, making
them vulnerable to physical tampering and unauthorized access.
 Vendor dependencies: OT systems often depend on third-party vendors for hardware,
software, and support, creating potential security risks if vendors do not implement robust
cybersecurity measures.
 Resource limitations: OT systems typically have limited processing power and memory,
restricting the ability to deploy advanced security features and tools.
 Lack of encryption: Data exchanged between OT devices is often not encrypted, leaving it
vulnerable to interception and manipulation by attackers.
 Data integrity concerns: Attacks that alter data, such as tampering with sensor readings,
can lead to critical decisions being made based on inaccurate information, causing severe
consequences.
OT Vulnerabilities
OT systems are increasingly integrated with IT networks, which has expanded their potential
attack surfaces. As OT and IT systems converge, the frequency of cyberattacks on IT networks
poses a risk to OT systems, as they may be compromised through IT infrastructure. Attackers
can exploit vulnerabilities in IT networks to launch attacks on OT systems. Table 18-08 covers
some common vulnerabilities in OT:

Vulnerability Description

1. Publicly Accessible OT  OT systems are linked directly to the internet, allowing


Systems third-party vendors to perform maintenance and
diagnostics remotely
 OT systems lack modern security measures to protect them
 Attackers can exploit vulnerabilities to perform password
brute-forcing or probe OT systems, potentially disabling or
disrupting their operations

2. Insecure Remote  Corporate networks utilize jump boxes to enable remote


Connections connections to the OT network
 Attackers can exploit weaknesses in jump boxes to gain
unauthorized remote access to OT systems
3. Missing Security  Using outdated software versions heightens risks and
Updates creates opportunities for attackers to breach OT systems
4. Weak Passwords  Operators and administrators often rely on default
usernames and passwords for OT systems, which are easy
to guess
 If the default vendor credentials for embedded devices and
management interfaces are not updated, attackers can gain
access to OT systems
5. Insecure Firewall  Incorrectly configured access rules permit unnecessary
Configuration connections between corporate IT and OT networks
 Support teams grant excessive access permissions to the
management interfaces on firewalls
 Vulnerable firewalls expose the OT network to security
risks, making them susceptible to attacks
6. OT Systems Placed  Corporate systems are linked to the OT network to access
within the Corporate IT operational data or transfer data to third-party
Network management systems
 OT systems like control stations and reporting servers are
within the IT network
 A compromised IT system can be used to access the OT
network
7. Insufficient Security  Attacks can also stem from OT systems, which rely on
for Corporate IT outdated legacy software and are accessible remotely
Networks from OT  Insecure OT devices can provide unauthorized access to
Systems corporate IT systems
8. Lack of Segmentation  Many OT networks are designed with a flat, unsegmented
within OT Networks structure, treating all systems as equally important and
functional
 The compromise of a single device can potentially expose
the entire OT network
9. Lack of Encryption and  OT networks' wireless devices rely on insecure and
Authentication for outdated security protocols
Wireless OT Networks  This vulnerability allows attackers to conduct sniffing and
authentication bypass attacks
10. Unrestricted  OT networks permit direct outbound connections for
Outbound Internet remote patching and maintenance tasks
Access from OT  Direct internet access to unpatched and insecure OT
Networks devices heightens the risk of malware attacks
 This setup makes OT networks vulnerable to malware and
command-and-control attacks
Table 18-08: OT Vulnerabilities

MITRE ATT&CK for ICS


The MITRE ATT&CK framework for ICS serves as a valuable resource for ICS security teams and
vendors. It enables them to analyze an attacker’s tactics against OT systems and devise effective
defense strategies. Additionally, it assists security teams in describing and analyzing an
attacker’s behavior following a compromise.
The MITRE ATT&CK framework for ICS outlines various tactics, each addressing specific
aspects of potential attacker strategies.
Initial Access
This tactic involves attackers' methods to establish their foothold within a targeted ICS
environment. Attackers may compromise OT assets, websites, IT resources, or other external
services to infiltrate the ICS network.
Techniques for Initial Access:
1. Drive-by Compromise
Attackers exploit vulnerabilities in a target user's web browser by tricking them into visiting a
compromised website during routine browsing, gaining unauthorized access to the OT system.
2. Exploiting Public-Facing Software
Attackers penetrate the OT network by exploiting known vulnerabilities in internet-facing
applications, often used for remote monitoring and management.
3. Exploiting Remote Services
Attackers leverage application vulnerabilities and error messages from operating systems,
programs, or kernels to conduct further attacks on remote services.
Additional Techniques for Initial Access:
● External remote services
● Internet-accessible devices
● Remote services
● Replication through removable media
● Rogue master
● Spear-phishing attachment
● Supply-chain compromise
● Transient cyber assets
● Wireless compromise
These techniques highlight attackers' diverse strategies to breach ICS environments and
underscore the need for robust security measures.
Execution
Execution involves an attacker’s efforts to run malicious code, alter data, or perform
unauthorized system functions using illegitimate methods. Various techniques are employed to
execute malicious actions within a device or asset in an ICT environment.
Techniques for Execution:
1. Changing the Operating Mode
Attackers manipulate the operating modes of a controller within the infrastructure (e.g.,
program download) to gain extended access to OT functionalities.
2. Command-Line Interface (CLI)
Attackers use the CLI to execute malicious commands and interact with the OT system. This
enables them to install and run harmful programs and carry out unauthorized operations while
evading detection.
3. Execution Through APIs
By injecting code into APIs, attackers can execute specific functions in a system when triggered
by the associated software.
Additional Techniques Used in the Execution Stage:
● Graphical User Interface (GUI)
● Hooking
● Modify controller tasking
● Native API
● Scripting
● User execution
These techniques demonstrate attackers' diverse methods to compromise systems and
emphasize the need for vigilant monitoring and preventive measures.
Persistence
Persistence refers to the technique attackers use to maintain their presence within an ICS
environment, even after device restarts or communication interruptions. These methods ensure
continued access and enable long-term attacks.
Techniques for Persistence:
1. Program Modification
Attackers alter or attach programs to a controller in an OT system, changing how it interacts
with other devices or processes within the environment.
2. Firmware Manipulation
Malicious firmware is embedded into hardware devices, allowing attackers to sustain access to
other systems and establish a foothold for prolonged attacks.
3. Project File Infection
Attackers inject harmful code into file dependencies, such as objects or variables essential for
operating Programmable Logic Controllers (PLCs). This often involves exploiting default PLC
functionalities.
Additional Techniques for Persistence:
● System firmware manipulation
● Use of valid accounts
These methods highlight how attackers secure their presence in ICS environments,
underscoring the need for robust defenses to prevent unauthorized access and minimize
vulnerabilities.
Privilege Escalation
Privilege escalation enables attackers to gain elevated access and permissions within an ICS
system or network, allowing them to carry out more advanced malicious activities.
Techniques for Privilege Escalation:
1. Exploiting Software Vulnerabilities
Attackers leverage known flaws or programming errors in software to increase their access
rights and privileges.
2. API Hooking
This technique involves intercepting and redirecting API calls within various processes,
enabling attackers to manipulate system behavior and elevate their privileges.
Evasion
Attackers employ evasion tactics to bypass standard defense mechanisms and avoid detection.
Techniques for Evasion:
1. Removing Indicators
Attackers eliminate evidence of their activities, such as attack indicators, from a compromised
host to obscure their presence and erase traces of the attack.
2. Rootkits
Rootkits are installed to conceal malicious activities by hiding services, connections, and system
drivers, effectively evading detection.
3. Changing the Operating Mode
By altering a controller’s operating mode, attackers can access additional system functionalities
while avoiding detection.
Additional Techniques for Evasion:
● Exploiting software vulnerabilities
● Masquerading
● Sending spoofed reporting messages
These techniques highlight attackers' various methods to remain undetected and evade security
measures.
Discovery
Discovery involves gathering information about an ICS environment to analyze and identify
potential target assets. Attackers use various techniques to collect data about the system's
infrastructure and operations.
Techniques for Discovery:
1. Enumerating Network Connections
Attackers analyze communication patterns between network devices to understand the
structure and functionality of the system.
2. Network Sniffing
This technique involves capturing or monitoring network traffic to extract critical details such
as protocols, source and destination addresses, and other key information.
3. Identifying Remote Systems
Attackers obtain information about other systems within the network, including hostnames, IP
addresses, and other identifiers, to facilitate malicious actions.
Additional Techniques for Discovery:
● Remote system information gathering
● Wireless network sniffing
These techniques demonstrate how attackers systematically collect intelligence to target and
exploit ICS environments.
Lateral Movement
Lateral movement involves attackers expanding their presence within the target ICS
environment by utilizing their existing access.
Techniques for Lateral Movement:
1. Exploiting Default Credentials
Attackers use built-in or factory-set credentials of control systems to carry out administrative-
level actions.
2. Program Downloads
Attackers can upload and execute user programs within a controller, gaining further control by
initiating program downloads.
3. Abusing Remote Services
By exploiting remote services, attackers can traverse the network, accessing additional assets
and components.
Additional Techniques for Lateral Movement:
● Exploiting vulnerabilities in remote services
● Transferring malicious tools laterally
● Using valid accounts for unauthorized access
These methods highlight how attackers propagate through the network to compromise more
systems and expand their influence within an ICS environment.
Collection
The collection encompasses attackers' methods for extracting data and acquiring insights into
the ICS infrastructure, including its domains and assets.
Techniques for Data Collection:
1. Automated Collection
Attackers utilize scripts or automated tools to gather information about the ICS environment
without systematic manual intervention.
2. Targeting Information Repositories
Sensitive details, such as control system layouts and specifications, can be extracted from
information repositories, giving attackers critical insights.
3. Accessing I/O Images
By obtaining the I/O image of a Programmable Logic Controller (PLC), attackers can access and
leverage memory for further malicious operations.
Additional Techniques for Data Collection:
● Detecting the operating mode
● Conducting man-in-the-middle attacks
● Monitoring process states
● Identifying points and tags
● Uploading programs
● Capturing screen data
● Wireless traffic sniffing
These strategies illustrate how attackers gather intelligence to understand and exploit ICS
environments effectively.
Command-and-Control
Command-and-control involves an attacker attempting to manipulate, disable, or exploit the
physical control processes within the targeted ICS environment.
Techniques for Command-and-Control:
1. Using Common Ports
Attackers exploit well-known ports like 80 (HTTP) and 443 (HTTPS) to communicate, thereby
avoiding detection by traditional security mechanisms.
2. Connection Proxy
A connection proxy allows attackers to redirect and control network traffic within the ICS
environment, facilitating covert operations.
3. Application-Layer Protocols
Attackers leverage standard application-layer protocols, such as HTTPS, telnet, and Remote
Desktop Protocol (RDP), to obscure their actions and gain control over targeted systems.
Inhibit Response Function
The inhibition of response function involves tactics where an attacker seeks to prevent or
disrupt actions taken in response to security incidents, such as hazards or failures.
Techniques for Inhibiting Response:
1. Activating Firmware Update Mode
Attackers may initiate firmware update mode to block normal security responses during a
security event, causing disruptions.
2. Blocking Command Messages
Attackers can intercept and block commands or instruction messages before they are delivered
to control systems, preventing appropriate responses.
3. Blocking Reporting Messages
By disrupting reporting messages from industrial systems, attackers can prevent them from
reaching their intended recipients, allowing their malicious activities to remain undetected.
Additional Techniques for Inhibiting Response:
● Alarm suppression
● Blocking serial communication (COM)
● Data destruction
● Denial-of-Service (DoS) attacks
● Restarting or shutting down devices
● Manipulating control I/O images
● Altering alarm settings
● Installing rootkits
● Stopping services
● Modifying system firmware
These techniques show how attackers can prevent systems from responding effectively to
security events, ensuring they can carry out malicious actions undetected.
Impair Process Control
Impairing process control involves attackers attempting to disable, manipulate, or take control
of the physical control processes within the targeted environment.
Techniques for Impairing Process Control:
1. I/O Brute-Forcing
Attackers may attempt to brute-force Input/Output (I/O) addresses to take control of a
process's functionality without directly targeting a specific interface.
2. Manipulating Parameters
By altering the control system's instruction parameters through programming, attackers can
disrupt or modify system operations.
3. Injecting Malicious Firmware
Attackers can inject malicious firmware into a device, reprogramming it to carry out harmful
tasks or enable further attacks.
Additional Techniques for Impairing Process Control:
● Sending spoofed reporting messages
● Issuing unauthorized command messages
These techniques illustrate how attackers can compromise process controls to disrupt or gain
control over operations in the target environment.
Impact
Impact refers to attackers' methods of inflicting damage, disrupting, or seizing control of the
data and systems within the targeted ICS environment and its related components.
Techniques for Impact:
1. Property Damage
Attackers can cause significant damage to both physical property and the surrounding
environment by launching various attacks on the ICS.
2. Loss of Availability
Attackers can make systems unresponsive to connected operations or communications by
disrupting or interfering with industrial processes.
3. Denial of Control
Attackers may manipulate control systems to sever communication between operators and the
process controls, leading to operational disruptions.
Additional Techniques for Impact:
● Denial of view
● Loss of control
● Decreased productivity and revenue
● Loss of protection and safety
● Loss of system visibility
● Manipulation of control or view
● Theft of operational data
These techniques highlight how attackers can profoundly negatively affect ICS environments'
operations, safety, and security.
OT Threats
With the merging of Operational Technology (OT) and Information Technology (IT), OT
systems are increasingly being used for functions they were not originally intended for. These
systems are now being integrated with IT networks and exposed to the global internet, creating
new vulnerabilities. Many OT systems still rely on outdated, legacy software lacking proper
security measures, offering a potential entry point for cybercriminals to infiltrate corporate IT
networks and OT infrastructure. Furthermore, the interconnected nature of OT networks,
which link machines and production systems, has led to more sophisticated cyberattacks that
can cause physical damage. Below are some key threats faced by OT networks:
1. Maintenance and Administrative Threats
Attackers target maintenance and administrative functions by exploiting zero-day
vulnerabilities within the OT network. By taking advantage of these vulnerabilities,
cybercriminals can inject and spread malware through IT systems and target interconnected
industrial control systems, such as SCADA and PLC.
2. Data Leakage
Attackers may exploit IT systems connected to the OT network to access the IT/OT gateway
and steal sensitive operational data, including configuration files, that are crucial for the
system's functioning.
3. Protocol Exploitation
Many OT systems rely on outdated legacy protocols and interfaces like Modbus and CAN bus,
often due to compatibility issues. Attackers take advantage of these protocols to launch attacks
on OT systems. For instance, they may exploit the emergency stop (e-stop) safety mechanism
to halt machinery in emergencies and carry out single-packet attacks.
4. Potential Destruction of ICS Resources
Attackers exploit weaknesses in OT systems to disrupt or degrade the functionality of ICS
infrastructure, potentially leading to critical safety and operational issues.
5. Reconnaissance Attacks
Many OT systems allow remote communication with little encryption or authentication,
enabling attackers to conduct initial reconnaissance and scanning. This helps them collect the
information needed for the later stages of the attack.
6. Denial-of-Service Attacks
Attackers use communication protocols like the Common Industrial Protocol (CIP) to launch
DoS attacks on OT systems. For instance, an attacker might send a malicious CIP connection
request to a device and, upon establishing the connection, inject a false IP configuration. If the
device accepts the configuration, it can cause a loss of communication between the device and
other connected systems.
7. HMI-Based Attacks
Human-Machine Interfaces (HMIs), sometimes called Hacker-Machine Interfaces, are
susceptible to exploitation due to existing vulnerabilities in OT systems. Despite advancements
in OT automation, human interaction with operational processes still presents security
challenges. The absence of global standards for HMI software development and insufficient
security measures often lead to vulnerabilities. Attackers can exploit these weaknesses to
conduct attacks such as memory corruption, code injection, and privilege escalation on OT
systems.
8. Exploiting Enterprise-Specific Systems and Tools
Attackers may target ICS components like Safety Instrumented Systems (SIS) by injecting
malware through exploited protocols. This allows them to detect hardware and systems used
for communication, further damaging or disrupting services.
9. Spear Phishing
Attackers send fraudulent emails containing malicious links or attachments that appear to come
from legitimate or trusted sources. When the victim interacts with the malicious content,
malware is injected, causing damage and spreading to other systems. For example, an attacker
may send an email with a harmful attachment to a system managing sales software in an
operational plant. Once the victim opens the attachment, the malware infects the sales software,
spreads to other networked systems, and damages industrial automation components.
10. Malware Attacks
Attackers are repurposing legacy malware that was initially designed to target IT systems,
adapting it to exploit OT systems. They conduct reconnaissance to find vulnerabilities in newly
integrated OT systems. Once identified, they use older malware versions to execute various
attacks. In some cases, attackers also develop new malware specifically designed to target OT
systems, including ICS/SCADA.
11. Exploiting Unpatched Vulnerabilities
Attackers exploit unpatched vulnerabilities in ICS products, firmware, and software used within
OT networks. ICS manufacturers design products for reliability and real-time performance but
often neglect built-in security features. Additionally, these vendors cannot produce patches for
vulnerabilities as quickly as IT vendors. This gap allows attackers to target and exploit these
vulnerabilities to attack OT systems.
12. Side-Channel Attacks
Attackers utilize side-channel attacks to extract sensitive information from OT systems by
analyzing their physical implementations. Techniques like timing and power analysis are
employed to gather critical data through these side-channel methods.
13. Buffer Overflow Attack
Attackers exploit buffer overflow vulnerabilities present in ICS software—such as those in HMI
web interfaces, ICS web clients, and communication interfaces—to inject malicious data and
commands. This manipulation alters the normal functioning and behavior of the systems.
14. Exploiting RF Remote Controllers
OT networks often rely on RF technology to remotely control industrial processes. However, RF
communication protocols generally lack built-in security, leaving them vulnerable to
exploitation. Attackers can exploit these weaknesses to disrupt industrial machines, gaining
unauthorized control or sabotaging production systems.
HMI-Based Attacks
Attackers often target HMI systems because they serve as the central control point for critical
infrastructure. If attackers gain control of an HMI, they can damage SCADA devices or gather
sensitive data about the system's architecture, which can be used for further malicious actions.
With this information, attackers may also disable alerts for potential threats to the SCADA
systems. Below are some SCADA vulnerabilities that attackers exploit to carry out HMI-based
attacks on industrial control systems:
 Memory Corruption
This type of vulnerability involves code security issues such as out-of-bounds read/write errors
and buffer overflows in the heap and stack. Memory corruption in an HMI occurs when the
memory contents are modified due to coding errors. If the altered memory is used, it can cause
the program to crash or behave in unintended ways. Attackers can trigger memory corruption
by overwriting the code to create a buffer overflow. In some cases, an unflushed stack may allow
attackers to manipulate the program using string manipulation techniques.
 Credential Management
This category's vulnerabilities include using hard-coded passwords, storing credentials in
unprotected formats like plaintext, and inadequately protecting credentials. Attackers can
exploit these weaknesses to gain administrative access to systems, allowing them to modify
system databases or configuration settings.
 Lack of Authorization/Authentication and Insecure Defaults
This category includes vulnerabilities such as transmitting sensitive data in plaintext, using
insecure default settings, ignoring encryption, and using insecure ActiveX controls for scripting.
In some SCADA systems, an administrator with legitimate access can view other users'
passwords. Attackers can exploit these weaknesses to gain unauthorized access to the system,
allowing them to record or alter the data being transmitted or stored.
 Code Injection
Vulnerabilities in this category involve common forms of code injection, such as SQL, OS,
command, and domain-specific injections. Gamma, a domain-specific language used for
Human-Machine Interfaces (HMIs), is particularly vulnerable to code injection attacks. This
scripting language creates fast-paced user interfaces and control applications. A vulnerability in
Gamma’s EvalExpression function, which evaluates, compiles, and executes code at runtime,
can be exploited by attackers to send and execute arbitrary commands or scripts on the SCADA
system.
 Buffer Overflow Vulnerabilities
Certain HMI software may be susceptible to buffer overflow vulnerabilities. Excessive data input
can overflow the allocated buffer, potentially allowing an attacker to execute arbitrary code on
the system.
 Path Traversal
Path traversal flaws in HMI web servers allow attackers to access files and directories outside
the designated web root folder, leading to unauthorized disclosure or manipulation of
information.
Side-Channel Attacks
Attackers carry out side-channel attacks by analyzing the physical implementation of a target
system to extract sensitive information. Two primary techniques used for these attacks are
timing analysis and power analysis. In timing analysis, attackers measure the time it takes for a
device to perform different computations. In power analysis, attackers focus on fluctuations in
power usage during cryptographic operations. ICS systems are particularly vulnerable to these
types of attacks.
 Timing Analysis
In this attack, passwords are often transmitted through a serial channel. Attackers use a loop
strategy to guess each character of the password, testing one character at a time. If the first
character is correct, the loop proceeds to the next character. If it is incorrect, the loop stops.
Attackers can deduce which characters are correct by measuring the time taken for each
password authentication attempt. While timing-based attacks can be detected and blocked,
they pose a threat.
 Power Analysis
Power analysis attacks are harder to detect and can allow the targeted device to continue
functioning after being compromised. As a result, attackers often prefer this method over timing
analysis to obtain sensitive information. This technique involves monitoring the power
consumption of semiconductors during clock cycles. An oscilloscope tracks the time interval
between pulses, and the resulting power profile can reveal how data is being processed,
potentially exposing sensitive information.
For example, by analyzing the power profile, attackers can determine a correct password
character by comparing it with an incorrect one. The same method can be used to extract
cryptographic keys. Suppose attackers gain physical access to an unsecured or unsupervised
device. In that case, they can use an oscilloscope, specialized hardware, and analysis software to
retrieve cryptographic keys.
Once these keys are obtained, attackers can modify the configuration of the targeted devices.
Since these systems are commonly used to secure power grids, such configuration changes can
cause significant damage. Such alterations can disrupt the system’s operation or transmit
incorrect data to the operator. As these devices are often part of a distributed network
controlled by a centralized system, incorrect data from one device can affect large portions of
the OT network.

Figure 18-84: Illustration of Side-Channel Attack

Hacking Programmable Logic Controller (PLC)


PLCs are vulnerable to cyberattacks because they manage the physical processes of critical
infrastructures. Attackers can discover exposed PLCs on the internet using tools like Shodan.
Once compromised, these PLCs can create significant security risks for organizations. Hackers
can manipulate the integrity and availability of PLC systems by targeting pin control operations,
allowing them to execute attacks like payload disruptions and PLC rootkits.
PLC Rootkit Attack

Figure 18-85: Hacking PLC Through PLC Rootkit Attack

Here are the steps involved in executing a PLC rootkit attack:


● Step 1: The attacker gains legitimate access to the PLC device by injecting a rootkit. They
then launch a control-flow attack against the PLC runtime to guess the default password,
ultimately achieving root access to the system.
● Step 2: The attacker identifies the input and output modules and their memory locations to
overwrite the PLC’s input and output parameters.
● Step 3: After mapping the I/O pins and logic, the attacker manipulates the I/O initialization
process, taking full control over the PLC’s operations.
A PLC rootkit exploits architectural weaknesses in microprocessors, enabling it to bypass
modern detection methods. This attack allows the attacker to fully control the PLC’s input and
output processes by manipulating the I/O initialization sequence. It is also called a "PLC ghost
attack" and requires detailed knowledge of the PLC architecture.
The PLC’s CPU operates in two modes: programming mode, where the PLC can remotely
download code from any computer, and run mode, which executes the code. Once attackers
gain access to the PLC, they can upload malicious code to the device, which replaces the original
code. The attacker can then control the input and output processes, manipulating mechanical
devices and potentially causing operational damage or failure.
Evil PLC Attack
In an Evil PLC attack, the attacker seeks out vulnerable PLC devices that are exposed to the
internet, often using online tools, to target OT workstations and disrupt production operations.
These exposed devices usually lack proper security protections, making them easy targets for
unauthorized access and data manipulation. Once a vulnerable PLC is identified, the attacker
transforms it into an "Evil PLC" by altering its configuration and modifying its behavior and
logic by downloading malicious code.
The steps involved in an Evil PLC attack are as follows:
 Step 1: The attacker locates a vulnerable PLC using online search tools like Shodan or Censys
 Step 2: The attacker takes advantage of the PLC’s firmware vulnerabilities and modifies its
programming logic through download processes
 Step 3: With the compromised PLC, the attacker triggers upload actions on connected
workstations, allowing the execution of arbitrary code
Figure 18-86: Illustration of Evil PLC Attack Methodology

Hacking Industrial Systems Through RF Remote Controllers


Industrial machines are commonly controlled through remote controllers, which are widely
used across sectors like manufacturing, logistics, mining, and construction for automation or
machine operation. These systems typically use a transmitter (TX) and receiver (RX) to
communicate within a network. The transmitter sends radio signals (through button presses)
while the receiver responds to these signals. If security measures in remote-controlled devices
are inadequately implemented, they can create significant vulnerabilities in industrial systems.
Attackers positioned within the range of the target system can utilize a specialized radio
transceiver to design and transmit their own packets, potentially gaining unauthorized access
to the system and carrying out harmful actions. Below are some threats industrial systems face
through RF remote controllers:
 Replay Attack: Attackers capture the RF signals (commands) sent by an operator and replay
them to the target system, effectively gaining control over the system.
Figure 18-87: Replay Attack on Industrial Systems

 Command Injection: Attackers familiar with RF protocols can manipulate or inject their
own RF packets through reverse engineering techniques to gain full machine control. They
record and analyze commands, reverse-engineer additional commands used to control the
device, and inject these commands to disrupt the device’s normal operations.

Figure 18-88: Command Injection Attack on Industrial Systems

 Exploiting E-stop: Using the knowledge gained in the previous step, attackers can
repeatedly send emergency stop (e-stop) commands to the device, halting its operations and
causing a Denial-of-Service (DoS).
Figure 18-89: Abusing e-stop to Perform a DoS Attack

 Re-pairing with Malicious RF Controller: An attacker can take control of the original
remote controller by pairing a malicious RF controller, which appears legitimate. By sending
false pairing requests and capturing the command sequence, they can hijack the device’s
controller and launch attacks on the target machine.

Figure 18-90: Malicious Re-Pairing Attack on an Industrial Machine

 Malicious Reprogramming Attack: Attackers can inject malicious code into the firmware
of remote controllers, allowing them to maintain persistent and complete remote access to
the industrial system.
Figure 18-91: Malicious Reprogramming Attack on an Industrial Machine

OT Supply Chain Attacks


Supply chain attacks in Operational Technology (OT) involve compromising a company's
suppliers' hardware, software, or services, subsequently used to breach the organization’s OT
environment. These attacks can be especially harmful, as they leverage trusted relationships and
can remain unnoticed for long periods. Table 18-09 discusses some of the primary methods
attackers employ to carry out supply chain attacks in OT settings:

OT Supply Chain Attacks Description

Third-Party Software Attackers insert harmful code into trusted software updates,
Compromise which, when installed, create backdoors or introduce malicious
functionalities.
Hardware Manipulation They modify hardware components during the manufacturing
or distribution, embedding malicious firmware or chips that
activate once deployed in the target environment.
Service Provider Breach Attackers infiltrate the target organization's OT network by
compromising service providers, such as maintenance or
support contractors, using stolen credentials, remote access
tools, or insider knowledge.
Injection of Malicious They tamper with the supply chain during shipping by
Components introducing malicious components or firmware or replacing
legitimate parts with compromised ones.
Exploitation of Trusted Attackers leverage the trust and access granted to suppliers,
Relationships subcontractors, or partners to move laterally within the target
organization's network.
Table 18-09: OT Supply Chain Attacks

OT Malware
Attackers are creating increasingly advanced malware to target industrial systems. Recent
examples like Fuxnet and INDUSTROYER.V2 have caused significant disruptions to operations
within industrial networks. This type of malware can potentially damage the software and
hardware that control critical infrastructure. Sometimes, OT malware can spread throughout a
network, rendering connected devices inoperable. Industrial control systems are particularly
vulnerable to such attacks due to their extensive network connectivity and the use of proprietary
systems and legacy technology that often go without regular updates or patches. OT
ransomware can lock or encrypt hard drive files, rendering the system inaccessible and
unusable. Here are a few notable examples of OT malware:
Fuxnet
Fuxnet is malicious software designed to disrupt Industrial Control Systems (ICS), which
manage critical infrastructure such as safety systems. Once deployed, it can manipulate
important data, block access to sensor gateways, and damage physical sensors. The malware can
overwrite the NAND chip in the gateways, disabling remote access and blocking administrative
control. Fuxnet can also exploit weak device credentials to gain root access to sensor gateways
and corrupt them, causing inaccurate readings and even physical damage to the equipment.

Figure 18-92: Deployment Scripts of the Fuxnet Malware


Figure 18-93: Fuxnet Showing Information About All Compromised Sensors

Other examples of OT malware include:


● Kapeka
● Abyss Locker
● AvosLocker
● COSMICENERGY
● INDUSTROYER.V2
● Pipedream
OT Malware Analysis: COSMICENERGY
COSMICENERGY is a new type of malware targeted at Operational Technology (OT) and
Industrial Control Systems (ICS). It is specifically designed to interfere with power operations
by interacting with IEC 60870-5-104 (IEC-104) devices, including Remote Terminal Units
(RTUs). This malware functions similarly to earlier variants like INDUSTROYER and
INDUSTROYER.V2, which are recognized for their capability to disrupt power grids via the IEC-
104 protocol.
Figure 18-94: Illustration of COSMICENERGY Infection Flow

COSMICENERGY Attack Scenario:


● Stage 1: Reconnaissance
Attackers typically begin by performing reconnaissance to pinpoint potential targets and collect
data about the target environment. This phase involves identifying MSSQL servers with access
to the OT network, obtaining MSSQL credentials, and locating IP addresses for target IEC-104
devices. Additionally, attackers may gather details about key assets, such as power-line switches,
circuit breakers within an RTU, or relay configurations.
● Stage 2: Initial Access
Attackers establish their initial foothold in the target environment by exploiting security flaws
or utilizing stolen credentials. For COSMICENERGY, this could involve connecting to a remote
MSSQL server provided by the user. To accomplish its goals, COSMICENERGY may upload and
execute the following files on the target system:

Filename Description Hash

MD5:
cd8f394652db3d0376ba24a990403d20
SHA1:
PIEHOP PyInstaller
r3_iec104_control.exe bc07686b422aa0dd01c87ccf557863ee6
executable
2f6a435 SHA256:
358f0f8c23acea82c5f75d6a2de37b6bea
7785ed0e32c41109c217c48bf16010
PIEHOP Python MD5:
r3_iec104_control compiled bytecode f716b30fc3d71d5e8678cc6b81811db4
entry point SHA1:
e91e4df49afa628fba1691b7c668af64ed
6b0e1d SHA256:
7dc25602983f7c5c3c4e81eeb1f2426587
b6c1dc6627f20d51007beac840ea2b
MD5:
c018c54eff8fd0b9be50b5d419d80f21
Decompiled PIEHOP SHA1:
r3_iec104_control.py entry point Python 4d7c4bc20e8c392ede2cb0cef787fe0072
script 65973b SHA256:
8933477e82202de97fb41f4cbbe6af325
96cec70b5b47da022046981c01506a7
MD5:
adfa40d44a58e1bc909abca444f7f616
SHA1:
PIEHOP Python
iec104_mssql_lib.pyc a9b5b16769f604947b9d8262841aa3082
compiled bytecode
f7d71a2 SHA256:
182d6f5821a04028fe4b603984b4d3357
4b7824105142b722e318717a688969e
MD5:
2b86adb6afdfa9216ef8ec2ff4fd2558
SHA1:
Decompiled PIEHOP
iec104_mssql_lib.py 20c9c04a6f8b95d2f0ce596dac226d56be
Python script
519571 SHA256:
90d96bb2aa2414a0262d38cc80512277
6a9405efece70beeebf3f0bcfc364c2d
MD5:
7b6678a1c0000344f4faf975c0cfc43d
SHA1:
LIGHTWORK
OT_T855_IEC104_GR.exe 6eceb78acd1066294d72fe86ed57bf43b
executable
c6de6eb SHA256:
740e0d2fba550308344b2fb0e5ecfebdd
09329bdcfaa909d3357ad4fe5552532
Table 18-10: COSMICENERGY Executable Files

● Stage 3: Executing the Attack


After gaining access to the target environment, the attackers initiate the COSMICENERGY
malware. COSMICENERGY consists of two primary components: PIEHOP and LIGHTWORK.
PIEHOP, developed in Python, connects with the MSSQL server to upload files and sends
remote commands to the RTU.

Figure 18-95: PIEHOP Command-Line Example


The entry point of PIEHOP (r3_iec104_control.py) calls the main function of PIEHOP, passing
the argument control=True. Following this, the r3_iec104_control.py file imports the
"iec104_mssql_lib" module.

Figure 18-96: PIEHOP Decompiled Entry Point


Figure 18-97: PIEHOP Main Function

LIGHTWORK (OT_T855_IEC104_GR.exe), developed in C++, generates IEC-104 ASDU


messages to modify the state of RTU Information Object Addresses (IOAs), switching them to
ON or OFF. This action impacts the operation of power-line switches and circuit breakers.
LIGHTWORK takes the following command-line arguments:
<ip_address> <port> <command> [either ON (1) or OFF (0)]

Figure 18-98: LIGHTWORK Command-Line Example

When OT_T855_IEC104_GR.exe is executed, LIGHTWORK sends a "C_IC_NA_1 – station


interrogation command" to the target station to retrieve its status. It then issues a "C_SC_NA_1
– single command" for each predefined Information Object Address (IOA) to toggle the state of
the target station’s IOA (either OFF or ON). Lastly, it sends a "C_CS_NA_1 – clock
synchronization command" to align the target station’s time with the time the device issues the
commands.

Figure 18-99: LIGHTWORK Traffic Captured in the NetworkMiner

If all the commands are executed successfully without any errors, LIGHTWORK displays the
following output on the command-line:

Figure 18-100: LIGHTWORK Results

As explained earlier, COSMICENERGY utilizes these two components to frequently disrupt


power by interacting with IEC-104 devices. By sending ON or OFF commands to these devices,
attackers can interrupt the electricity flow, potentially leading to large-scale outages and
damaging the power grid.
● Stage 4: Lateral Movement & Erasing Evidence
COSMICENERGY does not have typical lateral movement features, as it cannot spread across a
network. However, internal reconnaissance is needed to collect the necessary information for
execution, suggesting that attackers might have moved laterally within the network to gather
this data.
Once the attackers accomplish their goals, they try to erase all traces of malware and other
malicious actions from the target environment. This may involve removing the executable files
of PIEHOP and LIGHTWORK from the compromised systems.
OT Hacking Methodology
Operational Technology (OT) systems, including ICS/SCADA and DCS, are primarily designed
to oversee and manage industrial processes. These systems play a critical role in gathering data
related to physical parameters like temperature, pressure, valve settings, and input from human
operators while controlling devices such as electrical, hydraulic, mechanical, and pneumatic
actuators. Historically, OT systems and networks were completely separated from the internet.
However, growing business demands and the need for interoperability have led to the
integration of OT and IT networks. However, this integration has exposed OT systems to
vulnerabilities in IT networks, creating opportunities for cybercriminals to execute disruptive
attacks. This section delves into the methodology of hacking OT systems. It demonstrates the
use of various automated tools for such activities.
What is OT Hacking?
Today's industrial systems are more interconnected with the internet, making them increasingly
susceptible to vulnerabilities and cyber threats. Cybercriminals are employing advanced,
targeted attacks that can physically damage these systems. Organizations sometimes rely on
outdated software for compatibility purposes or share sensitive information with third parties
for remote equipment maintenance, further heightening security risks. OT hacking primarily
aims to disrupt or sabotage business operations by compromising industrial control systems at
manufacturing facilities. The integration of IT with OT has introduced numerous
vulnerabilities, such as remote sensors, Wi-Fi-enabled controllers, USB devices for software or
firmware updates, and cloud-based solutions like SCADA-as-a-Service. This interconnectedness
has made OT systems a prime target for cyberattacks.
How might a hacker exploit a compromised OT system?
 Gain full control of the system to cause damage or extract sensitive operational or
business data
 Shut down production lines or entire facilities, initiate DDoS attacks, and inflict
financial losses or tarnish a company's reputation
 Alter assembly processes to bypass essential production steps, leading to defective
products
 Manipulate industrial machinery to create safety hazards, such as overheating or
triggering emergency shutdowns, potentially endangering employees
 Deploy malware to disrupt operations within critical infrastructure
 Use ransomware to lock OT systems, demanding payment for restoring access
OT Hacking Methodology
The following steps outline the typical methodology for hacking an OT network:
● Information Gathering
● Vulnerability Scanning
● Launch Attacks
● Gain Remote Access
● Maintain Access
Information Gathering
The initial phase of compromising an OT network involves collecting detailed information
about the target systems and network using footprinting and reconnaissance methods. These
approaches enable attackers to map the network, identify devices connected to it, determine
the geographical location of the devices, retrieve default passwords for connected devices, and
detect open ports along with active services. Attackers commonly use tools like Shodan and
Nmap to gather critical data about the target OT network.
Identifying ICS/SCADA Systems using Shodan
The Shodan search engine serves as a valuable resource for attackers seeking information about
OT devices connected to the internet. This tool can uncover details about SCADA systems used
in critical infrastructure, such as water treatment facilities, nuclear power plants, HVAC
systems, electrical grids, and residential heating systems.
● Identifying SCADA Systems via Port Numbers
ICS/SCADA systems rely on various protocols specific to PLC manufacturers. Key SCADA
protocols include:
o Modbus: Port 502
o Fieldbus: Ports 1089-1091
o DNP: Port 19999
o Ethernet/IP: Port 2222
o DNP3: Port 20000
o PROFINET: Ports 34962-34964
o EtherCAT: Port 34980
By detecting these ports, attackers can pinpoint vulnerable SCADA systems accessible over the
internet. For instance, the query port:502 retrieves devices operating on Modbus port 502. It can
be used to search for ICS/SCADA systems with Modbus enabled.
● Identifying SCADA Systems Using PLC Information
Attackers can locate SCADA systems by searching for details such as PLC names, version
numbers, or manufacturer names. By leveraging Shodan, they can analyze system banners that
reveal key information, including the PLC name, manufacturer, and software versions.
For instance, Schneider Electric is a well-known ICS system provider that utilizes various
Modbus protocols. An attacker can use Shodan to search for system banners containing the
company's name to identify devices deploying Schneider Electric products.
Example Search:
The query "Schneider Electric" retrieves all systems associated with Schneider Electric products.
● Locating SCADA Systems by Geolocation
Attackers can also filter SCADA systems based on geographic location.
Example Search:
The query SCADA Country:"US" displays SCADA systems in the United States.

Figure 18-101: Shodan

Gathering Default Passwords using CIRT.net


The default password database provided by CIRT.net is an online resource that contains default
credentials for numerous devices, including those utilized in critical infrastructure.
Cybercriminals can exploit this database to access default passwords for various devices, such
as routers, switches, and Industrial Control Systems (ICS).
Figure 18-102: CIRT.net

Information Gathering Tools


Below are some tools used for gathering OT information:
Kamerka-GUI
Kamerka-GUI is a reconnaissance tool designed to identify and map Industrial Control Systems
(ICS) accessible via the internet. It leverages Shodan’s advanced search filters to locate specific
ICS devices, including SCADA systems, PLCs, HMIs, and RTUs. Using this tool, attackers can
target vulnerable ICS devices in specific countries. Kamerka-GUI provides an interactive map to
visualize the geographical locations of identified devices and can also generate heat maps to
highlight areas with the highest concentration of vulnerable systems.
Figure 18-103: Kamerka-GUI

Additional OT Information-Gathering Tools


Other tools for OT reconnaissance are discussed below:
● SearchDiggity (https://bishopfox.com)
● Zeek (https://zeek.org)
● Criminal IP (https://www.criminalip.io)
● ZoomEye (https://www.zoomeye.hk)
● ONYPHE (https://www.onyphe.io)
Scanning ICS/SCADA Systems using Nmap
Attackers often utilize scanning tools like Nmap to detect open ports and active services on
systems connected to OT networks. Below are some commonly used Nmap commands to
enumerate ports and services associated with ICS/SCADA systems:
● Identifying Open Ports and Services

nmap -Pn -sT --scan-delay 1s --max-parallelism 1 -p 80, 102, 443, 502, 530, 593, 789, 1089-
1091, 1911, 1962, 2222, 2404, 4000, 4840, 4843, 4911, 9600, 19999, 20000, 20547, 34962-34964,
34980, 44818, 46823, 46824, 55000-55003 <Target IP>
This command allows attackers to conduct initial reconnaissance by scanning for active
ICS/SCADA protocols. The ports specified in the command correspond to commonly used
ICS/SCADA protocol ports.
● Identifying HMI Systems

nmap -Pn -sT -p 46824 <Target IP>


Certain Human-Machine Interface (HMI) systems operate on ports outside the standard
ICS/SCADA range. For example, the HMI software Sielco Sistemi Winlog utilizes TCP port
46824. Attackers can use this command to identify HMI systems running on such ports.
● Scanning Siemens SIMATIC S7 PLCs

nmap -Pn -sT -p 102 --script=s7-info <Target IP>


This command identifies PLC devices with port 102 open. Siemens SIMATIC S7 PLCs use port
102 for S7 communication, facilitating data exchange between PLCs and SCADA systems.
● Scanning Modbus Devices

nmap -Pn -sT -p 502 --script modbus-discover <Target IP>


nmap -Pn -sT -p 502 --script modbus-discover --script-args='modbus-
discover.aggressive=true' <Target IP>
These commands detect Modbus-enabled devices and extract their Slave IDs, a key identifier
for such systems.
● Scanning BACnet Devices

nmap -Pn -sU -p 47808 --script bacnet-info <Target IP>


BACnet is used to manage building automation and HVAC systems. This command gathers
vendor name, device name, serial number, and firmware version data. It detects BACnet
Broadcast Management Devices (BBMDs) using the BACnet-discover-enumerate.nse script.
● Scanning Ethernet/IP Devices

nmap -Pn -sU -p 44818 --script enip-info <Target IP>


Ethernet/IP, a widely used industrial protocol, operates on UDP port 44818. The command
retrieves details such as vendor name, product code, device name, and IP address.
● Scanning Niagara Fox Devices

nmap -Pn -sT -p 1911,4911 --script fox-info <Target IP>


Niagara Fox, a protocol for device-to-device communication in Building Management Systems
(BMS), operates on TCP ports 1911 and 4911. The command reveals the application name, Java
version, host OS, time zone, local IP address, and software versions.
● Scanning ProConOS Devices

nmap -Pn -sT -p 20547 --script proconos-info <Target IP>


ProConOS is a high-performance runtime engine used in PLCs to control embedded and PC-
based systems. This command identifies PLC details, including type, project name, source code
name, and runtime ladder logic data.
● Scanning Omron PLC Devices

nmap -Pn -sT -p 9600 --script omron-info <Target IP>


nmap -Pn -sU -p 9600 --script omron-info <Target IP>
Omron's Factory Interface Network Service (FINS) protocol, which facilitates communication
and data exchange between PLC programs and remote devices, operates on TCP or UDP port
9600. These commands are used to identify Omron PLCs and gather related data.
● Scanning PCWorx Devices

nmap -Pn -sT -p 1962 --script pcworx-info <Target IP>


PCWorx provides PC-based automation solutions for industrial networks. These systems accept
unauthenticated messages from remote sources. The command helps attackers retrieve
information such as the PLC's type, model number, and firmware version.
Sniffing using NetworkMiner
NetworkMiner allows attackers to conduct passive network sniffing and packet capture to
identify open ports, hostnames, operating systems, sessions, and more without creating any
network traffic. Additionally, attackers use NetworkMiner to parse and analyze PCAP files,
reassembling or reconstructing transmitted files or certificates. Using NetworkMiner, attackers
can access and examine PCAP files to analyze previously captured network traffic from ICS
networks.

Figure 18-104: NetworkMiner

Analyzing Modbus/TCP Traffic using Wireshark


Wireshark is a free, open-source network protocol analyzer that allows for capturing and
examining network traffic. Cybercriminals can use this tool to monitor and analyze
Modbus/TCP traffic on industrial networks. They can inject a malicious payload into a Modbus
device by manipulating the captured packets. Since Modbus/TCP lacks built-in encryption or
security features, attackers can easily extract information from the data packets exchanged
between the network and the Modbus port on a device.
Figure 18-105: Wireshark

Discovering ICS/SCADA Network Protocols using Malcom


Malcolm is a robust network traffic analysis tool attackers can utilize to understand the
protocols used in Industrial Control Systems (ICS) environments. It offers comprehensive
visibility into network communications through two user-friendly interfaces: the OpenSearch
dashboard, a versatile data visualization plugin with various pre-built dashboards that provide
a clear overview of network protocols, and Arkime, a powerful tool designed to locate and
identify network sessions associated with potential security incidents. Furthermore, it ensures
secure communication through the user interface and remote log forwarders, employing
standard encryption protocols.
Figure 18-106: Malcolm

Vulnerability Scanning
After attackers collect information about a target OT network and systems, they scan
vulnerabilities to identify potential exploits and weaknesses in critical infrastructure and OT.
Tools like SCADA Family for Nessus and Skybox Vulnerability Control detect vulnerabilities in
OT and IT devices, protocols, and applications, including ICS/SCADA, PLCs, RTUs, HMIs,
gateways, desktop computers, and other connected systems. Additionally, attackers may use
tools like Wireshark to monitor and analyze industrial network traffic to uncover
vulnerabilities.
Vulnerability Scanning using Nessus
Nessus is a vulnerability assessment tool that enables attackers to identify weaknesses in ICS
and SCADA systems. It offers a quick overview of vulnerabilities linked to default policies and
templates, allowing attackers to create custom policies. Nessus helps attackers discover and
categorize vulnerabilities to launch attacks on targeted OT networks. It includes a range of
SCADA plugins that allow attackers to perform vulnerability scans on ICS/SCADA devices. The
vulnerabilities are identified based on the signatures of these plugins.
Steps to Perform Vulnerability Scanning on ICS/SCADA Systems Using Nessus
● Step 1: Log in to the Nessus web client using the credentials provided during installation.
Navigate to the Policies tab and select Create New Policy. Choose the Basic Network
Scan template.
Figure 18-107: Nessus Showing New Policy Settings

● Step 2: In the DISCOVERY node, adjust the settings for port scanning by specifying a port
range of 0–1000.

Figure 18-108: Nessus Showing New Policy Setting

● Step 3: Verify if SCADA plugins are available under the Plugins tab. If not, the results will
only show vulnerabilities related to non-SCADA ports.
Figure 18-109: Nessus Showing SCADA Plugins

● Step 4: Save the policy, open the My Scans folder, and select New Scan. Under the User
Defined policy section, choose the policy created in Step 1.

Figure 18-110: Nessus Showing Scan Templates

● Step 5: Select the policy and enter the required information in the provided fields, including
the target IP address, then click Launch.
Figure 18-111: Nessus BASIC Scan Settings

Upon completing the scan, the results will display discovered vulnerabilities. For example,
Nessus may identify two SCADA-related vulnerabilities, highlighted in yellow.

Figure 18-112: Nessus Showing Identified SCADA Vulnerabilities

After identifying the system vulnerabilities, attackers can exploit them using various techniques
to launch additional attacks on the targeted OT systems.
Vulnerability Scanning Using Skybox Vulnerability Control
Skybox performs comprehensive path analysis across integrated OT and IT networks, offering
insights into associated vulnerabilities and potential attack vectors. It can merge SCADA and
ICS data with information from attack vector analysis, intelligence feeds, SIEMs, and threat
intelligence sources. This tool allows for prioritizing millions of vulnerabilities in OT/IT
networks based on risk levels. Attackers can use Skybox to analyze and categorize network
vulnerabilities, facilitating the execution of various attacks on the IT/OT environment.

Figure 18-113: Skybox Vulnerability Control

Sniffing and Vulnerability-Scanning Tools


Sniffing Tool: SmartRF Packet Sniffer
The SmartRF Packet Sniffer comprises software and firmware designed to capture and display
over-the-air packets. This functionality is enabled by a capture device connected to a PC via
USB. It supports the CC13xx and CC26xx device families as capture devices and utilizes
Wireshark for packet display and filtering. The tool is compatible with protocols like ZigBee,
EasyLink, and BLE.
Figure 18-114: SmartRF Packet Sniffer

Vulnerability Scanning Tool: Microsoft Defender for IoT


Microsoft Defender for IoT conducts vulnerability assessments within IoT and ICS
environments, providing an objective risk score. It detects and identifies all IoT and ICS assets
connected to the network. The platform highlights device-level vulnerabilities, including
missing patches, weak passwords, unused open ports, and remote access ports. Additionally, it
generates reports on network-level vulnerabilities, such as unauthorized internet connections,
weak firewall configurations, rogue subnet links between IT, IoT, and ICS, unauthorized
Wireless Access Points (WAPs), and rogue devices.
Figure 18-115: Microsoft Defender for IoT

Fuzzing ICS Protocols


Fuzzing ICS protocols like Modbus, BACnet, and Internet Printing Protocol (IPP) is essential
for uncovering information and identifying critical network activities. Attackers leverage tools
such as Fuzzowski to detect potential errors and exploitable vulnerabilities in industrial
networks.
Fuzzowski
Fuzzowski is a network protocol fuzzer designed to assist attackers in performing fuzz tests on
ICS protocols. It supports the entire fuzzing process, including protocol fuzzing and
communication configuration. However, attackers must first thoroughly understand the
protocol they intend to fuzz.
Figure 18-116: Fuzzowski

Below are examples of fuzzing ICS protocols, including BACnet, Modbus, and IPP.
Fuzzing the BACnet protocol:
python -m fuzzowski 127.0.0.1 47808 -p udp -f bacnet -rt 0.5 -m BACnetMon

Figure 18-117: BACnet Monitor


Fuzzing Modbus:
python -m fuzzowski 127.0.0.1 502 -p tcp -f modbus -rt 1 -m modbusMon

Fuzzing IPP:
python -m fuzzowski printer1 631 -f ipp -r get_printer_attribs --restart smartplug

Figure 18-118: Fuzzing of IPP

Launch Attacks
During the vulnerability scanning, attackers identify weaknesses within the target industrial
network and systems. These vulnerabilities are subsequently exploited to carry out various
attacks, including HMI-based attacks, side-channel attacks, PLC exploitation, replay attacks,
and command injection attacks. Attackers commonly use tools like Metasploit and modbus-cli
to compromise PLC devices via the Modbus protocol.
Hacking ICS Hardware
Attackers leverage publicly available online resources to gather information about the hardware
chip used in a specific ICS device. This includes details such as the number of embedded pins,
connection configurations, and the supported types of I/O. They may also analyze the chip's
integrated software to extract data like certificates, key generation algorithms, and encryption
functions. With this information, attackers can manipulate analog and digital I/Os, alter the
device's normal operations, and reset or reboot processes.
Through static and dynamic analysis of the chip’s functions, attackers can identify arguments
used in functions and the presence or absence of I/O validations. This analysis often reveals
vulnerabilities like buffer overflows and other overlooked issues. Attackers can exploit these
vulnerabilities with various software and hardware tools to compromise ICS hardware and
further their malicious activities.
Below are examples of commonly used software and hardware tools attackers might utilize to
target ICS hardware:
Hardware Tools:
 Signal Analyzer: Used to test specific flags and analyze the binary operations of a chip’s
pins
 Multimeter/Voltage Meter: Employed to conduct tests similar to those performed by a
signal analyzer
 Microcontrollers and Memory Programmers: These tools help analyze and program
various chips, flash memories, and EPROMs
 Oscilloscope: Utilized to interpret analog or digital signals accurately
 Soldering Equipment: Used to attach or detach components like chips and memory,
allowing for isolated examination under controlled conditions
 Digital Microscope or Magnifying Glass: Enhances precision during soldering and aids
in reading small text or visualizing tiny device components
 Communication Interfaces (e.g., JTAG): Allow direct connection and communication
with ICS devices
 Screwdrivers and Precision Screwdrivers: Essential for disassembling devices to examine
their internal parts
 Precision Tweezers and Converters: Tools such as tweezers, UART converters, and USB
serial ports enable attackers to extract data directly from communication buses
Software Tools
 GDB: A debugging tool for Linux that helps attackers understand the execution processes
occurring on a chip
 OpenOCD: Facilitates connection between the attacker's system and the target chip for
analysis. Communication can be established using GDB on port 333 or through a telnet
interface via TCP port 4444
 Binwalk: This tool allows attackers to scan and analyze firmware binaries and images,
quickly identifying encryption methods, sizes, partitions, filesystems, and other relevant
details
 Fritzing: A design tool that aids attackers in creating electronic diagrams and circuit layouts
 Radare2: A portable framework attackers use for reverse engineering tasks, including
binary analysis
 Ghidra: A Software Reverse Engineering (SRE) platform that supports the analysis of
compiled code across Windows, macOS, and Linux. It offers disassembly, assembly,
decompilation, graphing, and scripting capabilities
 IDA Pro: A disassembler tool that converts machine-executable code into assembly
language, making it easier for attackers to interpret complex code
Hacking Modbus Slaves using Metasploit
Modbus masters and slaves communicate in plaintext without incorporating authentication
mechanisms, making the protocol vulnerable to exploitation. Attackers can take advantage of
this weakness by crafting and transmitting query packets resembling legitimate ones to Modbus
slaves, enabling them to access and manipulate the registers and coils of the slaves. This attack
requires the attacker’s machine to send packets formatted in the Modbus protocol to the
targeted slave. Attackers commonly use tools like Metasploit to execute various attacks on
Modbus slaves.
 Scanning Modbus Slaves: Attackers utilize the Metasploit module
auxiliary/scanner/scada/modbus_findunitid to identify and scan Modbus slaves
connected to the target's network, either within the Local Area Network (LAN) or via a
Modbus gateway.

Figure 18-119: Metasploit Scanning Modbus Slaves

 Manipulating Modbus Slave’s Data: Attackers employ the Metasploit module


auxiliary/scanner/scada/modbusclient to read from or write to the registers and coils of
the target Modbus slave.
Figure 18-120: Metasploit Reading Modbus Slave Registers

Figure 18-121: Metasploit Manipulating Modbus Slave Registers

Hacking PLC Using modbus-cli


PLCs are essential for managing industrial systems such as manufacturing plants, sewage
treatment facilities, power grids, and oil refineries. Attackers often target PLC devices like the
Schneider Electric TM221, which automates processes in various industries. These devices
communicate with other industrial equipment using the Modbus/TCP protocol. Attackers use
tools like modbus-cli to exploit PLC devices through the Modbus protocol.
Steps to Hack PLC Using modbus-cli:
 Step 1: Identify Internet-Connected PLCs
Tools such as Shodan and Nmap can locate industrial systems exposed to the internet. To find
Schneider Electric TM221 PLCs, search for "TM221ME16R" in the Shodan search bar. This will
display a list of Schneider Electric TM221 PLCs connected to the internet, many of which may
be vulnerable.
Figure 18-122: Shodan Showing Schneider Electric TM221 PLCs

 Step 2: Install modbus-cli


Once vulnerable PLC devices have been identified using Shodan, the next step is to install
modbus-cli by running the following command:
gem install modbus-cli
 Step 3: Understand Data Types
Before exploiting the PLC with modbus-cli, it is essential to understand the data types used to
read values. These data types rely on two addresses: Schneider and Modicon. A Schneider
address is denoted by %M followed by the address number.
Modicon
Datatype Data Size Schneider Address Parameter
Address
word (default, 16 bits %MW100 400101 --word
unsigned)
integer (signed) 16 bits %MW100 400101 --int
floating point 32 bits %MF100 400101 --float
double word 32 bits %MD100 400101 --dword
Boolean (coils) 1 bit %M100 101 N/A
Table 18-11: Modbus Data Types

 Step 4: Read Register Values


To read the register values from the identified devices in Step 1, use the following command:
For Schneider's address:
modbus read <Target IP> %MW100 10
For Modicon address:
modbus read <Target IP> 400101 10
These commands will retrieve ten words from the registers.

Figure 18-123: modbus-cli Reading Register Values

 Step 5: Manipulate Register Values


You can manipulate the register values by using the following commands:
modbus write <Target IP> %MW100 2 2 2 2 2 2 2 2
modbus write <Target IP> 400101 2 2 2 2 2 2 2 2
After executing these commands, the first eight register values will be replaced with 2.
 Step 6: Read Coil Values
To retrieve coil values, which are stored as Boolean data (ON/OFF or 1/0), use the following
commands:
modbus read <Target IP> 101 10
modbus read <Target IP> %M100 10
Figure 18-124: modbus-cli Reading Coil Values

 Step 7: Manipulate Coil Values


To manipulate the coil values using modbus-cli, use the following commands to turn on all the
coils:
modbus write <Target IP> 101 1 1 1 1 1 1 1 1 1 1
modbus write <Target IP> %M100 1 1 1 1 1 1 1 1 1 1
After executing these commands, if you check the coil values, all coils will show a value of 1.

Figure 18-125: modbus-cli Reading Coil Values

 Step 8: Capture Data to an Output File


To capture data from SCADA systems for later analysis and testing, use the following command
to record register values into an output file:
modbus read --output SCADAregisters.txt <Target IP> 400101 200
modbus read --output SCADAregisters.txt <Target IP> %MW100 200
To capture coil values into an output file, use:
modbus read --output SCADAcoils.txt <Target IP> 101 100
modbus read --output SCADAcoils.txt <Target IP> %M100 100
Figure 18-126: modbus-cli Capturing Data into the Output File

Gain and Maintain Remote Access


During the information gathering and vulnerability scanning stages, attackers analyze the OT
environment to identify weaknesses that could grant them remote access to industrial control
systems. For instance, they may exploit flaws in industrial protocols or deploy malware to
initiate attacks aimed at breaching these systems. Once attackers have infiltrated the industrial
infrastructure, they can alter the functionality and operations of control systems, leading to
both physical harm and financial losses for the organization. After gaining access, attackers
often use these compromised devices as a launching point for attacking other connected devices
within the network. To maintain control and facilitate further exploitation, attackers employ
tactics such as clearing logs, updating firmware, or embedding rootkits to remain undetected.
Once they have access to the target device, attackers can manipulate the firmware of devices
like PLCs to carry out firmware-based attacks that allow them to monitor and control various
processes.
Gaining Remote Access using DNP3
Internet connected control systems are commonly used across power plants, manufacturing,
and construction industries. These systems are designed to allow remote monitoring or control
of operations. However, they are often set up with direct internet access, bypassing firewall
protections or using default credentials for access. Attackers can exploit these vulnerabilities,
such as poorly configured networks or weak/default passwords, to gain unauthorized access to
industrial systems. Since default credentials are widely known and weak passwords can be easily
cracked through brute force, attackers can easily exploit these security weaknesses.
Attackers can utilize online tools like Shodan to scan for open ports or services on targeted ICS
devices. Once they identify an open port, they can exploit its vulnerabilities to gain remote
access to industrial systems.
For example, when targeting ICS protocols such as DNP3 on port 20000, attackers can use
Shodan to perform a port scan, which reveals open ports and their associated vulnerabilities. By
selecting an open port, attackers are directed to the system's login page, where they can attempt
to access the ICS network or systems using default credentials or brute-forcing the passwords.
Figure 18-127: Open Ports on Shodan
Figure 18-128: Shodan Showing Open Port DNP3

OT Hacking Tools
Attackers employ OT hacking tools to identify industrial control systems connected to the
target network, examine legacy software installed on these devices, and detect vulnerable ports,
services, and insecure communication protocols. These tools help launch attacks on the target
systems and network. This section covers different OT hacking tools.
Below are several tools commonly used by attackers to compromise OT systems and networks:

mbtget
mbtget is a command-line tool written in Perl that facilitates Modbus transactions. This tool
enables attackers to access the Modbus protocol's TCP and RTU versions through the MBclient
object, targeting ICS systems and networks.
Figure 18-129: mbtget

Additional tools for hacking OT systems and networks are listed below:
● CSET (https://github.com)
● Attkfinder (https://gitlab.com)
● ICSREF (https://github.com)
● ICSFuzz (https://github.com)
● ISF (https://github.com)

OT Attack Countermeasures
This section covers a range of OT security strategies, OT vulnerabilities and their corresponding
solutions, security measures aligned with the Purdue model, global OT security organizations,
and OT security solutions and tools. Organizations can implement effective mechanisms to
safeguard critical industrial infrastructure and related IT systems against cyber threats by
adopting these security measures.
How to Defend Against OT Hacking
To mitigate OT hacking risks, the following countermeasures are recommended:
 Regularly perform risk assessments to minimize exposure to potential threats
 Deploy specialized sensors designed to identify vulnerabilities within the network passively
 Utilize threat intelligence to detect potential risks and prioritize OT patching to protect
critical assets
 Consistently update OT hardware and software tools to maintain security
 Disable unused ports and services to reduce attack surfaces
 Apply secure configurations and follow secure coding practices for OT applications
 Ensure systems are upgraded to the latest technologies and patched frequently
 Maintain a comprehensive asset inventory to monitor outdated or unsupported systems
 Continuously analyze log data from OT systems for real-time attack detection
 Provide employees with training on updated security policies and enhance their awareness
of emerging threats
 Enforce strong, hashed passwords and replace default factory-set passwords
 Protect remote access by implementing multi-layered security measures such as two-factor
authentication, VPNs, encryption, and firewalls
 Develop and implement incident response and business continuity plans
 Strengthen the network perimeter to block and filter unauthorized inbound traffic
 Conduct regular scans of systems and networks using anti-malware tools
 Limit network traffic using techniques such as rate-limiting and whitelisting to mitigate
DDoS and brute-force attacks
 Enhance system security by disabling unnecessary services and functions
 Apply patches for vulnerabilities promptly, as released by manufacturers
 Monitor DNS logs frequently to identify unauthorized access attempts
 Update and secure systems interacting with ICS/SCADA devices to prevent exploitation that
could bypass security gateways
 Engage professional red teams to identify vulnerabilities in critical industrial infrastructure
 Deploy Intrusion Detection Systems (IDS) and flow-monitoring tools to detect potential
attacks early
 Validate and sanitize all input data to prevent attacks like buffer overflow, command
injection, and Cross-Site Scripting (XSS)
 Use library functions instead of external processes to achieve desired functionality securely
 Process SQL queries using prepared statements, parameterized queries, or stored
procedures in ICS systems
 Host ICS web applications only on tested and trusted third-party web servers
 Design ICS systems to restrict unauthorized access and enforce the principle of least
privilege
 Verify the integrity of transmitted messages by appending checksums to each message
 Ensure ICS vendors incorporate cryptographic signatures into application updates
 Conduct periodic audits of industrial systems to validate security controls, production, and
management practices
 Use Demilitarized Zone (DMZ) connections between ICS and corporate networks to enable
secure communication
 Validate the bounds and integrity of network data on server applications processing ICS
protocol traffic
 Review source code for ICS applications handling network traffic to identify potential
vulnerabilities
 Implement network traffic monitoring tools with Deep Packet Inspection (DPI) capabilities
to detect malicious activities in OT networks
 Deploy Next-Generation Firewalls (NGFWs) with DPI features to monitor and filter traffic
at network perimeters
 Utilize OT-specific Intrusion Detection and Prevention Systems (IDPS) designed to detect
and block cyber threats targeting industrial control systems
 Integrate encryption, authentication, and integrity verification mechanisms to protect
industrial protocols like Modbus, DNP3, and OPC
 Equip OT devices with endpoint protection solutions to guard against malware,
ransomware, and other cyberattacks
 Establish secure remote access in OT environments through solutions like Virtual Private
Networks (VPNs) with Multi-Factor Authentication (MFA) and encrypted sessions
 Implement Public Key Infrastructure (PKI) to authenticate and encrypt communications
between servers, PLCs, engineering workstations, and clients
OT Vulnerabilities and Solutions
Industrial systems like ICS/SCADA, PLCs, and RTUs are vulnerable to cyber threats, posing
substantial risks to critical infrastructure. To safeguard these systems, organizations must
implement effective security measures and controls. The following table outlines some of the
most prevalent OT vulnerabilities along with their corresponding solutions:
Vulnerability Solutions

1. Publicly Accessible OT Systems  Enable multi-factor authentication for enhanced security


 Utilize enterprise-level firewalls and secure remote access
solutions
 Create and routinely test incident response plans to ensure
preparedness
2. Insecure Remote Connections  Employ a secure multi-factor authentication system and
enforce stringent password policies
 Adopt effective security patch management practices
 Utilize Role-Based Access Control (RBAC) to regulate
remote access permissions
3. Missing Security Updates  Conduct application testing in a sandbox environment
before deploying them live
 Use firewalls and apply device hardening measures
4. Weak Passwords  Establish distinct username conventions for corporate IT
and OT networks
 Replace default credentials during installation
 Conduct security audits to ensure compliance with secure
password policies across both IT and OT networks
5. Insecure Firewall Configuration  Set up a secure firewall configuration
 Define and manage Access Control Lists (ACLs) on the
firewall
6. OT Systems Placed within the  Separate corporate IT and OT devices
Corporate IT Network  Create a Demilitarized Zone (DMZ) for all IT and OT
systems connections
 Continuously monitor the DMZ
7. Insufficient Security for  Limit access to the IT/OT network according to business
Corporate IT Networks from OT requirements
Systems  Set up a secure gateway between the OT and IT networks
 Conduct regular risk assessments
8. Lack of Segmentation within  Differentiate between critical and non-critical systems
OT Networks  Implement a zoning model that applies a defense-in-depth
strategy
 Adopt a zero-trust security model, assuming no trust by
default
9. Lack of Encryption and  Implement robust wireless encryption protocols
Authentication for Wireless OT  Utilize industry-standard cryptographic algorithms
Networks  Perform regular security audits
10. Unrestricted Outbound  Perform a formal risk assessment
Internet Access from OT  Carefully monitor and isolate OT systems from external
Networks access
 Store security updates in a separate repository outside the
OT network
Table 18-12: OT Vulnerabilities and Solutions

How to Secure an IT/OT Environment


The convergence of IT and OT is increasingly embraced by industries such as traffic control
systems, power plants, and manufacturing companies. These IT/OT systems are frequently
targeted by attackers seeking to exploit vulnerabilities and launch cyberattacks. According to
the Purdue model, the IT/OT environment is structured into multiple levels, with each level
requiring appropriate security measures.
Table 18-13 outlines various types of attacks on different Purdue levels of an IT/OT environment,
the associated risks, and the security controls needed to strengthen the network against cyber
threats:
Security
Zone Purdue Level Attack Vector Risks
Controls
Enterprise 5 & 4 Spear phishing, Abusing Firewalls, IPS,
(Enterprise Ransomware infrastructure, Anti-bot, URL
Network and Access to the filtering, SSL
Business network inspection,
Logistics Antivirus, DLP
Systems)
Industrial DMZ 3.5 (IDMZ) DoS attacks Malware Anti-DoS
injections, solutions, IPS,
Network Antibot,
infections Application
control, ALF
Manufacturing 3 (Operational Ransomware, Altering Anti-bot, IPS,
Systems) Bot infection, industrial Sandboxing,
Unsecured USB processes, Application
ports Industrial Control, Traffic
spying, encryption, Port
Unpatched protection
monitoring
systems
Manufacturing 2 & 1 (Control DDoS Altering IPS, Firewall,
Systems and exploitation, industrial Communication
Basic Controls) Unencrypted processes, encryption using
protocols, Industrial IPsec, Security
Default spying gateways, Use of
credentials, authorized RTU
Application and and PLC
OS commands
vulnerabilities
Manufacturing 0 (Physical Physical security Modifications or Point-to-point
process) breach disruptions to communication,
the physical MAC
process authentication,
Additional
security
gateways at
levels 1 and 0
Table 18-13: OT Vulnerabilities and Solutions

Implementing a Zero-Trust Model for ICS/SCADA


Attackers aiming to disrupt ICS infrastructure are increasingly targeting OT networks. To stay
ahead of such threats, organizations must adopt effective security controls and address
vulnerabilities proactively to prevent advanced attacks. Many ICS networks rely on outdated
systems or hardware lacking modern security features and access controls, making them more
susceptible to sophisticated attacks. By implementing a zero-trust model, organizations can
enhance access management for legacy systems and networks, ensuring thorough visibility and
validating all applications, users, and devices within the ICS network.
Steps to Implement a Zero-Trust Model in an ICS Network:
 Step 1: Defining the Network
As an organization's attack surface continuously evolves, securing the entire organization
becomes challenging. The zero-trust approach should begin by defining the attack surface,
which involves identifying the organization’s assets, sensitive data, and critical applications
within control centers or factory floors.
 Step 2: Mapping the Traffic
It is essential to map and document network traffic flow to understand how network devices
and resources interact. Traffic mapping helps OT teams gain full visibility into the network,
allowing them to determine the necessary security measures to protect critical data and
applications.
 Step 3: Designing the Network
Once the traffic flows are understood, security analysts can design a Zero-Trust Architecture
(ZTA) tailored to the organization’s business needs. This can begin with deploying a Next-
Generation Firewall (NGFW), which introduces a segmentation gateway to protect critical
areas. This setup enhances access control and enables internal evaluations of the protected
surface.
 Step 4: Establishing a ZT Policy
A Zero-Trust (ZT) policy should be created to whitelist users and devices following the
network's architecture design. This allows security analysts to specify who can access the ICS
network, why they need access when they can, and which resources are available.
 Step 5: Ongoing Monitoring and Maintenance
In the final step, security analysts must ensure that the Zero-Trust Architecture (ZTA) can
monitor network traffic as planned, providing valuable insights into the network. They should
also manage updates for all network devices as needed to maintain security and efficiency.

International OT Security Organizations


As OT systems become more widespread and integrated with IT, security experts must exercise
heightened caution and implement robust security policies to safeguard OT networks. Several
global cybersecurity organizations are dedicated to providing security frameworks and insights
to enhance the resilience of critical infrastructure.
Below are a few international organizations that notify companies about potential threats and
offer IT/OT solutions to defend OT industries against cyberattacks:
OTCC
The Operational Technology Cybersecurity Coalition (OTCC) is an industry-driven initiative to
enhance the security of Operational Technology (OT) environments. With OT systems such as
Industrial Control Systems (ICS), SCADA systems, and other critical infrastructures increasingly
targeted by cyber threats, the OTCC works to tackle these issues through collaboration,
information sharing, and the development of best practices.

Figure 18-130: OTCC

OT-ISAC
The Operational Technology Information Sharing and Analysis Center (OT-ISAC) is a central
hub for sharing threat intelligence within OT industries, including energy and water utilities.
The organization provides various tools and methods to securely exchange information between
the OT/IT spectrum to protect industrial systems and networks from cyber threats. Through its
connections with multiple information-sharing centers, OT-ISAC gathers data on emerging
threats and offers timely solutions to strengthen the industrial systems of its member
companies.
Figure 18-131: OT-ISAC

NERC
The North American Electric Reliability Corporation (NERC) is a non-profit international
regulatory body focused on ensuring the reliable and secure operation of the electric grid. NERC
establishes and enforces reliability standards, conducts annual assessments of seasonal and
long-term reliability, monitors the bulk power system for awareness, and provides education,
training, and certification for industry professionals.
Figure 18-132: NERC

Industrial Internet Security Framework (IISF)


The Industrial Internet Security Framework (IISF) focuses on mitigating the risks posed by
unforeseen threats from both internal and external sources that could disrupt production. The
primary goal of this framework is to identify and monitor the integration of IT and OT
operations while prioritizing potential threats.
Figure 18-133: Industrial Internet Security Framework (IISF)

ISA/IEC-62443
The ISA/IEC-62443, developed by the International Society of Automation (ISA) and the
International Electrotechnical Commission (IEC), is a non-profit professional organization
representing engineers, technicians, and managers in industrial automation. This standard
offers a flexible framework to address and reduce current and future security risks in Industrial
Automation and Control Systems (IACS), which are integral to the OT sector. It also outlines
technical cybersecurity requirements for various IACS components, such as embedded devices,
network elements, host systems, and software applications, focusing on OT industries. The
standard defines the security features that allow components to defend against threats at
specific security levels without relying on additional compensating measures for OT systems.
Figure 18-134: ISA/IEC-62443

OT Security Solutions
The industrial and corporate sectors are increasingly digitizing their operational processes,
expanding access to OT devices across a wider range of the internet. However, the costs
associated with managing security in heavy industries are often underestimated, resulting in
various security challenges. To mitigate these risks, industries should invest in cybersecurity
programs and solutions.
Cybersecurity professionals should carefully assess the current challenges and requirements,
adapting security measures to meet these evolving needs while implementing necessary
operational changes. Consequently, many established OEM providers and start-ups have
introduced new strategies and technologies to safeguard the OT environment.
Given the decentralized nature of heavy industries, security solutions can be integrated into all
technology-related decisions within both IT and OT. Information Risk Management (IRM) can
also serve as a second line of defense, with some industries even implementing internal audits
as a third line of defense.
Some of the emerging technology solutions organizations are using to safeguard the OT
environment include:
 Firewalls
Firewalls are crucial for monitoring and controlling network traffic, enhancing security by
inspecting traffic between OT and IT networks. They can identify and block new threats,
preventing attackers from moving between networks after compromising a system. It is
recommended that critical assets be placed in a DMZ separate from SCADA systems. Tools like
FortiGate Rugged Next-Generation Firewalls and OTIFYD Next-Gen OT Firewalls are
commonly used.
 Unified Identity and OT Access Management
Access management centralizes operations such as adding, securing, modifying, and removing
user access to OT systems, integrated with the organization’s identity-management system for
robust authentication. This approach minimizes risk by granting limited privileges to superuser
accounts, helping security personnel track critical assets, and identifying attack sources. Tools
like Claroty and MetaDefender IT-OT Access are used to manage and secure access to industrial
systems.
 Asset Inventory and Device Authorization
Asset inventory tools connect only authorized devices to the OT network, detecting all
connected devices and their vulnerabilities based on manufacturer, version, and type. These
tools also identify device faults and improve device efficiency. Tools such as SCADAfence,
OTbase, Guardian, and Dragos are used for asset inventory and device authorization.
 OT Network Monitoring and Anomaly Detection
OT network monitoring tools continuously track industrial systems and traffic in a noninvasive
manner, detecting anomalies—unexpected or malicious events—using machine-learning
algorithms. These tools help quickly identify malicious behaviors. Security professionals can use
tools like iSID and Rhebo OT Security for monitoring and anomaly detection.
 Decoys to Mislead Attackers
Decoys, or honeypots, employ deception technology in the OT environment to automatically
set up traps that attract attackers, helping expose their presence and actions. This provides an
additional layer of defense against those attempting to infiltrate the industrial network.
Security experts can utilize tools like Attivo Networks ThreatDefend, Conpot, and GasPot to
enhance network protection.

EXAM TIP: Understand the security measures that can be implemented to protect OT
environments. This includes recognizing best practices for securing devices and networks
against cyber threats.

OT Security Tools
Here are some tools available for securing OT systems and networks:
Flowmon
Flowmon helps manufacturers and utility companies maintain the reliability of their industrial
networks, preventing downtime and service disruptions. It achieves this through continuous
monitoring and anomaly detection, enabling the rapid identification and resolution of issues
such as malfunctioning devices, cyber espionage, zero-day vulnerabilities, or malware.
Figure 18-135: Flowmon

Here are some other tools for securing an OT environment:


● Tenable OT Security (https://www.tenable.com)
● Nozomi Networks (https://www.nozominetworks.com)
● Forescout (https://www.forescout.com)
● FortiGuard (https://www.fortinet.com)
● RAM2 (https://www.otorio.com)

EXAM TIP: Pay attention to vulnerabilities commonly found in OT systems, such as weak
authentication, lack of encryption in communication, and absence of regular updates or patches.

Summary
In this chapter, we explored IoT concepts, along with various IoT technologies and protocols.
We also examined the different threats and attacks targeting IoT networks and devices.
Additionally, we covered the IoT hacking methodology, which includes information gathering,
vulnerability scanning, launching attacks, gaining remote access, and maintaining access. The
chapter also highlighted various IoT hacking tools. Furthermore, we discussed several
countermeasures to prevent IoT network hacking attempts by threat actors and provided
insights into securing IoT networks and devices using dedicated security tools.
We also delved into OT concepts, the associated threats and attacks, and the OT hacking
methodology and tools. Various countermeasures to defend against OT attacks were discussed,
and the chapter concluded with a demonstration of OT security solutions and tools.
Mind Map

Figure 18-136: Mind Map

Practice Questions
1. Which of the following is a key characteristic of IoT?
A. Limited connectivity
B. Intelligent sensors
C. Manual data processing
D. Static design

2. In the IoT architecture, which layer is responsible for initial data processing and as a bridge
between devices and clients?
A. Application Layer
B. Middleware Layer
C. Access Gateway Layer
D. Edge Technology Layer

3. Which protocols are specifically designed for short-range communication in IoT devices?
A. MQTT
B. LoRaWAN
C. ZigBee
D. 6LoWPAN

4. What distinguishes the Device-to-Cloud communication model in IoT?


A. Devices connect directly to each other via Bluetooth or ZigBee.
B. Devices communicate with a gateway for protocol conversion.
C. Devices exchange data with the cloud instead of directly with the client.
D. Devices use NFC for communication within a limited range.

5. Which challenges make IoT devices particularly vulnerable to cyberattacks?


A. High power consumption of IoT devices.
B. Lack of robust security policies.
C. Limited range of communication protocols.
D. Difficulty in hardware production.

6. Which of the following is NOT a primary threat category associated with IoT devices?
A. Security
B. Privacy
C. Interoperability
D. Safety

7. Which of the following OWASP top 10 IoT threats involves exploiting weak or absent
authentication in web portals or APIs?
A. Insecure Network Services
B. Insecure Ecosystem Interfaces
C. Lack of Secure Update Mechanisms
D. Use of Insecure or Outdated Components

8. What is the primary risk associated with a DDoS attack targeting IoT devices?
A. Unauthorized physical access to devices.
B. Tampering with firmware during updates.
C. Overloading target systems to disrupt services.
D. Interception of sensitive data during transmission.

9. What is the primary goal of a Distributed Denial-of-Service (DDoS) attack?


A. To steal sensitive data from the target system.
B. To inject malware into IoT devices.
C. To overwhelm a target system with traffic, rendering it inaccessible.
D. To compromise HVAC systems remotely.

10. In a Rolling Code Attack, what technique does the attacker use to exploit a vehicle's
locking mechanism?
A. Injecting malware into the vehicle's system.
B. Using a jamming device to block and intercept rolling codes.
C. Gaining remote access through default credentials.
D. Overloading the car's communication system with traffic.

11. What is a key characteristic of a BlueBorne attack?


A. It targets Bluetooth-enabled devices and requires active user interaction.
B. It exploits vulnerabilities in IoT systems like HVAC systems.
C. It compromises Bluetooth-enabled devices without user interaction or pairing.
D. It manipulates signal transmission using Software-Defined Radio (SDR).

12. What is the primary objective of a Sybil attack in Vehicular Ad-hoc Networks (VANETs)?
A. To gain unauthorized physical access to vehicles.
B. To create the illusion of traffic congestion using fake identities.
C. To steal sensitive data from vehicle sensors.
D. To disable vehicle communication modules.

13. Which type of attack involves capturing valid communication messages and resending
them to disrupt IoT devices?
A. Exploit kit attack
B. Side-channel attack
C. Replay attack
D. Man-in-the-middle attack

14. What makes exploit kits particularly dangerous in IoT environments?


A. Their ability to mimic legitimate devices.
B. Their ability to exploit cryptographic weaknesses.
C. Their adaptability to new vulnerabilities.
D. Their reliance on physical access to devices.
15. Which of the following is a characteristic of a side-channel attack?
A. Using fake identities to manipulate network traffic.
B. Substituting legitimate devices with forged ones.
C. Monitoring power consumption or electromagnetic emissions to extract encryption keys.
D. Sending malicious requests to IoT devices to gain control.

16. Which tool is designed to provide detailed information about internet-connected devices,
such as routers and CCTV cameras?
A. MultiPing
B. Shodan
C. FOFA
D. Wireshark

17. Which of the following is a key feature of the FCC ID Search tool in IoT device information
gathering?
A. Identifying device vulnerabilities in real-time.
B. Searching for devices based on their IP address.
C. Retrieving detailed device certifications and test reports.
D. Intercepting network traffic for unauthorized access.

18. Which tool is designed for real-time packet capture in Z-Wave networks?
A. Nmap
B. Suphacap
C. IoTSeeker
D. Genzai

19. What does the IoTSeeker tool primarily focus on during vulnerability scanning?
A. Scanning IPv6 capabilities of IoT devices.
B. Detecting IoT devices with factory-set credentials.
C. Analyzing the radio spectrum of IoT devices.
D. Conducting buffer overflow vulnerability tests.

20. Which tool performs spectrum analysis for radio communication in IoT devices?
A. ONEKEY
B. Gqrx
C. beSTORM
D. IoTVAS

21. Which of the following is a hardware device used for SDR-based attacks on IoT devices?
A. GNU Radio
B. ChipWhisperer
C. RTL-SDR
D. ONEKEY

22. What is the primary purpose of ChipWhisperer in IoT security research?


A. Capturing and analyzing radio signals.
B. Conducting side-channel power analysis and glitching attacks.
C. Monitoring network traffic and endpoints.
D. Performing spectrum analysis.

23. Which tool generates Software-Defined Radio (SDR) signals?


A. GNU Radio
B. ChipWhisperer
C. IoTSeeker
D. beSTORM

24. What is the primary function of a Distributed Control System (DCS)?


A. Automating machinery without human intervention.
B. Supervising multiple local controllers and controlling distributed field devices.
C. Acquiring real-time data for centralized control and monitoring.
D. Protecting industrial environments during hazardous events.

25. What distinguishes a Safety Instrumented System (SIS) from a Basic Process Control
System (BPCS)?
A. SIS is primarily used for automating complex industrial processes.
B. SIS includes safety interlocks to prevent equipment damage.
C. SIS ensures safety by transitioning processes to a predefined safe state during hazardous
events.
D. SIS generates production data reports for operational optimization.
Answers
1. Answer: B
Explanation: Intelligent sensors are a fundamental characteristic of IoT. They enable devices
to detect, collect, and process data from their environment, making them crucial for the
functionality and efficiency of IoT systems.
2. Answer: C
Explanation: The Access Gateway Layer facilitates the initial data processing and bridges the
communication between devices and clients. It manages functions like message routing,
identification, and subscriptions, ensuring seamless data transfer within the IoT system.
3. Answer: C
Explanation: ZigBee is a protocol designed for short-range, low-power communication. It is
commonly used in home automation systems and devices requiring infrequent data
transmission, making it ideal for IoT applications.
4. Answer: C
Explanation: In the Device-to-Cloud communication model, devices rely on cloud services for
data exchange and command transmission, often using Wi-Fi or Ethernet. For example, a Wi-
Fi-enabled CCTV camera uploads its feed to the cloud, which can be accessed by authorized
users.
5. Answer: B
Explanation: IoT devices often lack robust security measures, making them susceptible to
cyberattacks. To mitigate these risks, manufacturers must integrate security measures from the
design phase through deployment and maintenance.
6. Answer: C
Explanation: The primary IoT threat categories are security, privacy, and safety, as they directly
relate to the vulnerabilities and risks posed by IoT devices. Interoperability, while important, is
not considered a direct threat category.
7. Answer: B
Explanation: Insecure ecosystem interfaces, such as web portals and APIs, are vulnerable to
attacks due to weak or missing authentication, poor encryption, and insufficient validation. This
poses a risk to the security of IoT devices.
8. Answer: C
Explanation: A Distributed Denial-of-Service (DDoS) attack overwhelms target systems,
rendering them incapable of delivering services. This attack typically uses a network of
compromised IoT devices (botnets).
9. Answer: C
Explanation: The main objective of a DDoS attack is to flood a target system with excessive
traffic from multiple sources, making it slow, unresponsive, or completely inaccessible to
legitimate users.
10. Answer: B
Explanation: In a Rolling Code Attack, the attacker uses a jamming device to block the signal
from the key fob to the vehicle while intercepting the transmitted rolling codes. These codes
are later replayed to unlock the vehicle.
11. Answer: C
Explanation: A BlueBorne attack takes advantage of vulnerabilities in the Bluetooth protocol,
allowing attackers to infiltrate nearby devices without requiring user interaction, prior pairing,
or device discovery mode.
12. Answer: B
Explanation: In a Sybil attack, an attacker uses multiple fake identities to create the illusion of
traffic congestion, disrupting the communication between nodes in a VANET. This can severely
impact network performance and lead to misinformation about traffic conditions.
13. Answer: C
Explanation: In a replay attack, attackers capture valid communication messages and
repeatedly send them to the target device. Denial-of-Service (DoS) attacks, message
manipulation, or even device crashes can result in denial-of-service.
14. Answer: C
Explanation: Exploit kits are malicious tools that automatically adapt to new vulnerabilities by
incorporating updated exploitation methods and features. This allows them to remain effective
and undetected in IoT environments.
15. Answer: C
Explanation: In a side-channel attack, attackers exploit signals emitted by devices, such as
power usage or electromagnetic emissions, to gather information about encryption keys. This
indirect method of accessing sensitive data is quick and minimally invasive.
16. Answer: B
Explanation: Shodan is a specialized search engine that indexes information about internet-
connected devices. It helps attackers gather details like IP addresses, hostnames, and device
locations, making it an essential tool for reconnaissance during IoT hacking.
17. Answer: C
Explanation: The FCC ID Search tool allows users to access detailed information about a
device, including certifications, test reports, and manuals, by entering the device's unique FCC
ID. This data helps attackers identify potential vulnerabilities in the device.
18. Answer: B
Explanation: Suphacap is a hardware-based sniffer specifically designed to monitor and
capture data packets in Z-Wave networks. It supports various Z-Wave controllers, such as
SmartThings and Vera, making it ideal for real-time traffic capture.
19. Answer: B
Explanation: IoTSeeker scans networks for IoT devices still operating with default credentials,
such as admin passwords, making them susceptible to hijacking attacks. It focuses on
HTTP/HTTPS services to identify these vulnerabilities.
20. Answer: B
Explanation: Gqrx is a software-defined radio tool that analyzes the radio spectrum. It
monitors frequency bands used by devices like car key fobs and sensors, providing attackers or
researchers insights into IoT communications.
21. Answer: C
Explanation: RTL-SDR is a hardware device used as a USB dongle to capture radio signals in
the surrounding area without an internet connection. It supports a wide frequency range and
enables activities like decoding GPS signals and sniffing GSM signals.
22. Answer: B
Explanation: ChipWhisperer is designed for embedded hardware security research. It focuses
on side-channel attacks, such as power analysis, to extract cryptographic keys and manipulate
hardware behavior through glitching techniques.
23. Answer: A
Explanation: GNU Radio is a software tool for generating SDR signals using external RF
hardware. It provides signal processing units for creating and analyzing SDR signals, making it
useful for SDR-based attacks.
24. Answer: B
Explanation: DCS is designed to manage production systems within a geographic location by
using centralized supervisory control to oversee distributed local controllers.
25. Answer: C
Explanation: SIS is designed to protect industrial environments by taking over when BPCS
fails, shutting down systems, or transitioning them to a safe state to prevent accidents.

You might also like