Implementing UFMC as a Candidate
Modulation Scheme for 5G Network in
MATLAB Program
Mazen Al-Ashabi Haron Al-Kahtani
Ammar Ajlan Mousa Abo-Taleb
Nabeel Moeed
Supervisor:
Dr. Jamil Sultan
Outline
Introduction OFDM Basics [comprise (FDM
Technique, Applications and
Problem Statement Advantages)]
Project's Objectives OFDM Drawbacks [comprise
4G and 5G Requirements (Why we need UFMC Technique)]
Multi-Carrier and Single-Carrier UFMC Basics [comprise
Systems (Applications and Advantages)]
Simple Compare Between OFDM
Inter-Symbol Interference (ISI) and UFMC Techniques
and Inter-Carrier Interference
(ICI) Problems Results Analysis and Discussions
Recommendations
Orthogonality Concept
1
Introduction
Problem Statement Project's Objectives
To meet higher data rates and improve To evaluate how UFMC fulfills the
access flexibility as the main given 5G requirements.
requirement for future radio To investigate how UFMC and
communication system 5G. ODFM techniques can be
Using UFMC as a candidate waveform implemented by using MATLAB
for 5G aims to overcome OFDM software.
shortcomings like high Out-of Band To investigate the properties of
(OOB) radiation and high sensitivity to UFMC waveforms.
timing and frequency offsets.
To know the PSD and BER of OFDM
Moreover, UFMC reduce the drawbacks and UFMC.
of FBMC such as the longer time delay To know the PAPR of OFDM and
of FBMC filters to filtrate the sub- UFMC.
carriers
2
4G and 5G Requirements
Latency Latency
20 ms 1 ms
MBB
eMBB
Always on Data Rate
Data Rate IoT
Multiple 10 Gbps
1Gbps M2M
4G 5G
Devices
Mobility Mobility
Massive
350 Km/h MIMO 500 Km/h
MIMO
100 Mbps < 1 Gbps
3
Single-Carrier and Multi-Carrier Systems
Single-Carrier Systems Multi-Carrier Systems
• Serial Transmission • Parallel Transmission
• EX: CDMA, QAM • EX: OFDM, UFMC
Spectrum Spectrum
f f
Single carrier
Carriers
4
Inter-Carrier Interference(ICI) and Inter-
Symbol Interference (ISI) Problems
Inter-Carrier Interference Inter-Symbol Interference
Symbol 1 Symbol 2
ICI ISI
Symbol 1 Symbol 2
Guard- band Guard- band
5
Orthogonality Concept
𝟐𝝅 𝟏 ∆f
∆𝑾 = or ∆𝒇 = Peak
𝑵 𝑻𝒔
Where
∆𝑾 or ∆𝒇 is the Carrier Spacing
N is the no. of FFT points
Ts is the Symbol Interval
𝒃
𝒇 𝒂 𝒕 × 𝒈 𝒕 𝒅𝒕 = 𝟎 null
6
OFDM Basics
FDM Technique OFDM vs FDM Techniques
Guard- band
Saving of
Bandwidth
7
OFDM Basics
OFDM Transmitter and Receiver Structure
CP
QAM IFFT
1100100101 P/S
S/P
.. N-Points
..
Modulation Channel
CP
1100100101 QAM FFT
P/S .. Equalizer .. N-Points .. S/P
Demodulation
8
OFDM Basics
Cyclic Prefix (CP) in OFDM
• Copy of last part of OFDM symbol
• To eliminate ISI OFDM OFDM
Symbol 2 Symbol 1
• Restore the Orthogonality
• CP > Delay Spread CP CP
• Wasting to the Spectrum
Identical Copy
9
OFDM Basics
OFDM Applications OFDM Advantages
• In Wire Scenarios • Resistant to ISI
ADSL and HDSL • Permits High Data Rate
PLC • Flexibility (Due to Implement the
IFFT / FFT)
• In Wireless Scenarios
IEEE 802.11 a/g/n/ac
LTE
DVB and DAB
WiMAX / IEEE 802.16
10
OFDM Basics
OFDM Drawbacks
• High Sensitive to Frequency Offset
• High PAPR
• OOB Radiation
11
UFMC Basics
UFMC Concept
• To overcome OFDM Drawbacks
• To Fulfill the 5G Requirements
• Multi-Carrier Waveform Technique
• Using Sub-Band Filter (Chebyshev Filter)
• Using Equalizer
12
UFMC Basics
UFMC Transmitter and Receiver Structure
Band
QAM IFFT
1100100101 P/S Filter
S/P
.. N-Points
.. Length L
Modulation Channel
FFT
1100100101 QAM
P/S .. Equalizer .. 2N- .. S/P
Points
Demodulation
ZP
13
UFMC Basics
UFMC Advantages
• Enabling Low Latency Modes
• Good Spectral Efficiency
• Suitable for 5G System
• Permits Higher Data Rate Than ODFM
• Reduces OOB Radiation
14
Simple Compare Between OFDM and UFMC
Techniques
OFDM Technique UFMC Technique
• Using CP • No Using CP
• No Using Filter • Using Filter
• High OOB Radiation • Low OOB Radiation
• High PAPR • Lower PAPR Than OFDM
• Higher Latency • Lower Latency Than OFDM
• High-Spectral Efficiency • Higher-Spectral Efficiency
• Doesn't Fulfill the 5G • Fulfill the 5G Requirements
Requirements
15
Results Analysis and Discussions
System Parameters of UFMC Technique
Parameters Parameters Description
Number of FFT points 512
Sub-band Size 20
numSub-bands 10
sub-bandOffset = numFFT/2-sub-bandSize *numSub-bands/2 for
156
band center
filterLen = similar to cyclic prefix length 43
2: 4-QAM, 4: 16-QAM, 6: 64-QAM, 8: 256-QAM bitsPerSub-Carrier = 6
side-lobe attenuation, dB 40
SNR in dB 15
16
Results Analysis and Discussions
UFMC Simulation Process
• Generation of Signals
17
OFDM Basics
UFMC Simulation Process
• QAM Modulation • The Output Signal From IFFT Process
18
Results Analysis and Discussions
UFMC Simulation Process
• Sub-Band Filter Output • Zero Padding
N + L – 1 = 512 + 43 -1
= 554 points.
470-point zero-padding
Channel
19
Results Analysis and Discussions
UFMC Simulation Process
• FFT Output
20
Results Analysis and Discussions
UFMC Simulation Process
• Select Data Sub-Carriers (After Remove • UFMC Symbols before the
Zero Padding and Down-sampling By 2) equalization process
21
Results Analysis and Discussions
UFMC Simulation Process
• Equalization
22
Results Analysis and Discussions
UFMC Simulation Process
• Equalization
23
Results Analysis and Discussions
UFMC Simulation Process
• QAM Demodulation
24
Results Analysis and Discussions
UFMC Results
• Power Spectral Density (PSD)
25
Results Analysis and Discussions
UFMC Results
• Out-of-Band (OOB)
UFMC has around 23 dB better
performance on the OOB
OOB = 340 - 317
340 dB
317 dB
Average power of all UFMC symbols Average power of all OFDM symbols
26
Results Analysis and Discussions
UFMC Results
• Bit Error Rate (BER) versus Signal to Noise Ratio (SNR)
Number of FFT points = 512
Modulation Scheme Bits per Symbol SNR (dB) UFMC BER OFDM BER
1 0.035 0.025
5 0.0025 0
QPSK 2
7 0 0
15 0 0
1 0.191 0.288
100 %
4 5 0.096 0.256
16-QAM 99.4 %
7 0.053 0.251
15 0.0015 0.24
93 %
1 0.2625 0.365
5 0.18 92.85 % 0.358
64-QAM 6
7 0.14083 0.351
27 15 0.024167 0.338
Results Analysis and Discussions
UFMC Results
• Bit Error Rate (BER) versus Signal to Noise Ratio (SNR)
Number of FFT points = 1024
Modulation Scheme Bits per Symbol SNR (dB) UFMC BER OFDM BER
1 0.0025 0.0025
5 0 0
QPSK 2
7 0 0
15 0 0
1 0.08625 0.2615
0%
5 0.0225 0.241
16-QAM 4 100 %
7 0.00375 0.2425
15 0 0.24
100 %
1 0.18583 0.3633
5 0.1025 99.9 % 0.35
64-QAM 6
7 0.0725 0.344
28 15 0.00083333 0.338
Results Analysis and Discussions
UFMC Results
• Peak to Average Power Ratio (PAPR)
number of FFT points = 512
Modulation UFMC OFDM
QPSK 9.04 dB 8.4377 dB
7.28 %
16-QAM 8.2379 dB 8.8843 dB
4.5 % 13 %
64-QAM 8.6229 dB 9.9269 dB
number of FFT points = 1024
Modulation UFMC OFDM
QPSK 7.7732 dB 8.4377 dB
6.6 %
16-QAM 8.2967 dB 8.8843 dB
14 % 3%
64-QAM 9.6202 dB 9.9269 dB
29
Recommendations
UFMC technique is more quite suitable for the future generation. UFMC will
be more reliable and efficient for next-generation wireless communication.
To reduce PAPR of UFMC, applying the PAPR reduction techniques like PTS,
SLM and optimization algorithms like Particle Swarm Optimization (PSO).
Implementing the UFMC system with Kaiser window can provide slightly
improved PAPR characteristics compared to UFMC with Dolph–Chebyshev
filter.
Using simulation programs like MATLAB and SYSTEMvue software will help
and contribute to enhancing the practical performance of the waveforms for
next-generation wireless communication
Search more about 5G waveforms especially about UFMC waveform and
whatever comes next.
30
Thank You
