Hindawi

Wireless Communications and Mobile Computing

Volume 2018, Article ID 8514314, 9 pages
[Link]

Research Article
NOMA for Multinumerology OFDM Systems

1 1,2
Ayman T. Abusabah and Huseyin Arslan
1
School of Engineering and Natural Sciences, Istanbul Medipol University, 34810 Istanbul, Turkey
2
Department of Electrical Engineering, University of South Florida, Tampa, FL 33620, USA

Correspondence should be addressed to Ayman T. Abusabah; asabah@[Link]

Received 23 November 2017; Revised 21 February 2018; Accepted 28 March 2018; Published 9 May 2018

Academic Editor: Oğuz Kucur

Copyright © 2018 Ayman T. Abusabah and Huseyin Arslan. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.

Nonorthogonal multiple access (NOMA) is a promising technique which outperforms the traditional multiple access schemes
in many aspects. It uses superposition coding (SC) to share the available resources among the users and adopts successive
interference cancelation (SIC) for multiuser detection (MUD). Detection is performed in power domain where fairness can
be supported through appropriate power allocation. Since power domain NOMA utilizes SC at the transmitter and SIC at the
receiver, users cannot achieve equal rates and experience higher interference. In this paper, a novel NOMA scheme is proposed for
multinumerology orthogonal frequency division multiplexing system, that is, different subcarrier spacings. The scheme uses the
nature of mixed numerology systems to reduce the constraints associated with the MUD operation. This scheme not only enhances
the fairness among the users but improves the bit error rate performance as well. Although the proposed scheme is less spectrally
efficient than conventional NOMA schemes, it is still more spectrally efficient than orthogonal multiple access schemes.

1. Introduction Power domain NOMA adopts multiplexing multiple

users through sharing the same resources at the transmitter
Cumulative and incessant demands on new services and (TX). Furthermore, it uses successive interference cancela-
applications, in addition to the great expansion in the number tion (SIC) as a multiuser detection technique to separate
of connected devices, have led to a huge data traffic explosion the users through power differences at the receiver (RX)
and appearance of different application classes and classifi- side. NOMA is considered as a promising multiple access
cations [1]. Thus, a strong need to boost the expected high technique for the fifth-generation wireless networks due to
data traffic has been recently emerged. Also, since it became massive connectivity, low latency, and high spectral efficiency
obvious that next generation has to support high data rates, (SE) [4].
applications industry and academia agreed on the necessity It is shown that, by adopting NOMA with orthogonal
of new and flexible radio access technologies (RATs) [2]. frequency division multiplexing- (OFDM-) based method
Multiple access (MA) techniques play a major role in the as RAT, multiple users can be allocated on a subcarrier at
overall communication process. Since the first generation up the same time. However, the cochannel interference (CCI)
until the fourth generation, MA techniques were distributing per subcarrier increases as more users are multiplexed on
the users over the available resources orthogonally. For the same subcarrier, which degrades the system performance
instance, time division multiple access, frequency division [5]. In [6], the authors have studied users’ pairing; then they
multiple access, code division multiple access, and orthogonal concluded that NOMA outperforms OMA especially with
frequency division multiple access (OFDMA) are considered users whose channel conditions are more distinctive, which
as orthogonal multiple access (OMA) techniques. Eventually, is practically difficult to occur.
researchers have moved away from utilizing the resources in Power control/allocation has been studied in many works
an orthogonal way to a nonorthogonal one, which they call [7]. To guarantee the fairness among NOMA users, more
nonorthogonal multiple access (NOMA) [3]. power is required for users with poor channel conditions
2 Wireless Communications and Mobile Computing

FFT FFT
N 2N user 2
1 1

0.8 0.8
user 3 user 3

0.6 user 2 0.6

Power

Power
0.4 user 1 0.4 user 1

0.2 0.2

0 0
−4 −2 0 2 −4 −2 0 2
Normalized Bandwidth Normalized Bandwidth
(a) Carrier representation for conventional NOMA at the RX end (b) Carrier representation for proposed NOMA at the RX end

user 3
user 2
user 1 user 3 user 1
Frequency

T
BＮＩＮ CP + Frequency user 2
BＮＩＮ CP +

T 2T
Time Time
(c) Time representation for conventional NOMA at the TX end (d) Time representation for proposed NOMA at the TX end

Figure 1: Signals superposition in time and frequency domains.

and less power for users with better channel conditions [8]. Even though the narrow subcarriers share the bandwidth
However, if the users have similar channel conditions, OMA with the wide subcarriers, the detection, of wide subcarrier
can guarantee better fairness and conventional power domain users, is independent of narrow subcarrier users; therefore,
NOMA cannot strictly guarantee the users’ quality of service SIC can be used to detect the wide subcarrier users based
(QoS) targets [9], which could be critical for some scenarios on their power differences. On the other hand, the narrow
with strict fairness constraints. subcarriers are detected once the interference imposed by
In this work, we propose an OFDM based NOMA wide subcarriers is eliminated.
scheme. The scheme utilizes the numerology concept to By assigning one of the users narrower subcarriers, we
reduce the constraints associated with the conventional reduce the amount of CCI imposed on the wide subcarrier
NOMA schemes. Simply, the users utilize different subcarrier users. Furthermore, the interference imposed on the narrow
spacings, wide subcarriers, and narrow subcarriers. A three- subcarrier user can be easily canceled by eliminating the wide
user scenario is considered in this paper. In the conventional subcarrier signals, which enhances the bit error rate (BER)
scheme, the three users are supposed to share the same performance for each user.
subcarriers as shown in Figure 1(a), while in the proposed The narrow subcarrier user has an extra degree of free-
scheme one user is assigned narrower and less frequently dom (DOF) as its power level is independent of the detection
spaced subcarriers as illustrated in Figure 1(b). process; that is, it is not restricted to the wide subcarrier
The subcarriers configuration of the proposed scheme is users and does not affect their detection process. In other
characterized by the fact that the wide subcarrier users are words, the SIC process does not depend on the power level
fully overlapped within the same wide subcarriers and make of the narrow subcarriers which grants more flexibility for
a zero crossing at the peaks of the other wide subcarriers. power assigning. Based on that, the proposed scheme has the
Furthermore, the narrow subcarriers do not impose any advantage of providing a fairer rate allocation to the users
interference at the peaks of wide subcarriers, that is, by compared with the conventional scheme especially when the
avoiding the transmission on the half of narrow subcarriers. channel conditions of the users are similar.
As a result, the wide subcarriers are not affected by any The subcarriers configuration of Figure 1(b) can be simply
external interference. On the other hand, an interference is accomplished by composing the symbols of wide subcarrier
imposed by the tails of the wide subcarriers on the peaks of users synchronously in the time domain, that is, each with
the narrow subcarriers. a length of one OFDM symbol slot. Then, the extended
Wireless Communications and Mobile Computing 3

symbol, that is, narrow subcarrier user, is added. However, other at the TX side. However, processing them together at
a novel structure is adopted for the transmission. As shown the RX side, that is, by using an extended FFT window, makes
in Figure 1(d), the OFDM symbol slots of wide subcarrier them overlapping and therefore nonorthogonal. As the basic
users are orthogonally constructed and then the extended NOMA concept based on multiplexing the users at the TX
symbol of narrow subcarrier user is added. In this case, utilizing the same resources, the proposed scheme can be
the wide subcarrier users are composed at the RX end and considered as a half NOMA.
the proposed subcarrier configuration is obtained. In both The key advantage of using larger FFT window size at RX
transmission structures, the resulting signal consists of two is the capability of equalizing the channel using one CP for
OFDM symbol slots and the utilized resources are the same. the whole OFDM symbols and thus increasing the SE. This
To achieve this purpose, a fast Fourier transform (FFT) property can be even used for a pure OFDMA system. For
operation, with the length of two OFDM symbol slots, is example, in the absence of user 2 in Figure 1(d), the system
performed at the RX end. By doing this, the wide subcarrier becomes an OFDMA system. Then, if FFT operation, with
users are multiplexed at the RX end. Extending the length of the length of two OFDM symbols, is performed at the RX, the
FFT window at the RX end allows us to equalize the channel transmitted symbols are composed and one CP can be used
using one cyclic prefix (CP) [10], which is a good solution to for equalization rather than two. In our case, the composed
increase the SE even with an absolute OFDMA system. signals are detected in power domain and user 2 is able to
The rest of the paper is organized as follows: Section 2 share the other users’ resources without introducing any extra
presents the signal configuration of the proposed scheme. interference.
Problem description, features, and potentials for our tech- Note that user 2 signal (narrow subcarriers) makes a zero
nique are provided in Section 3. The system model for con- crossing with the other signals (wide subcarriers) at the peaks.
ventional multicarrier NOMA scheme is given in Section 4. This can be clearly concluded from frequency domain repre-
The design analysis is provided in Section 5. Simulation sentation. The same result is not obvious from time domain
results are shown in Section 6. Finally, Section 7 concludes representation. In Section 5, it is proven mathematically that,
our paper. by adopting an extended FFT window size at the RX end,
the wide subcarrier users are multiplexed although they are
2. Signal Configuration assigned different OFDM symbol slots at the TX and the
narrow subcarrier users do not affect their detection process.
The frequency and time representations, for conventional
NOMA scheme, are shown in Figures 1(a) and 1(c), respec- 3. Problem Description
tively. Three different power signals are multiplexed utilizing
the same resources, where each signal constitutes one OFDM To describe the features of our proposed scheme, two scenar-
symbol slot; then a CP is appended. Therefore, the SE is ios are considered. The first scenario appears in Figure 2(a)
expected to be doubled 3 times compared with OFDMA where three downlink users have distinctive channel condi-
system without considering the CP redundancy. tions |ℎ|, while the second scenario appears in Figure 2(b),
On the other hand, in proposed NOMA, one of the users where two users have similar channel conditions; that is, two
(user-2) is assigned narrower and less frequent subcarriers as users are at the cell edge and one user is close to the base
illustrated in Figure 1(b), which simply multiplies the symbol station (BS).
duration by two as depicted in Figure 1(d). Since the new In the case of the users whose channel conditions are
structure uses only 𝑇 seconds out of a possible 2𝑇, the SE, similar, like the second scenario, either we assign similar
defined as bps/Hz, is halved for the wide subcarrier users. power allocation (PA) to achieve the fairness and, therefore,
Meanwhile, as the total energy budget is the same, the power SIC cannot work properly due to its inherent nature which
used by wide subcarrier users is twice as before. As the SE depends on power differences for separation, or we assign
increases logarithmically with power, this increase does not different PA which leads explicitly to unfairness distribution.
compensate for the .5 loss in the SE; thus, the overall SE
improvement over OFDMA is greater than 1 but strictly less 3.1. Conventional NOMA (CN). If users have distinctive |ℎ|
than 1.5. as represented in Figure 2(a), under perfect channel state
As mentioned earlier, the subcarriers’ configuration in information assumption at the BS, achieving the fairness can
Figure 1(b) can be obtained if both symbols of user 1 and be ensured through a proper PA. Furthermore, degradation
user 3 share the first half of user 2 symbol. However, in in the performance due to the number of assigned users is
Figure 1(d), user 1 and user 3 are constructed orthogonally expected. For instance, user 3 signal has to be detected with
at the TX. In this case, the subcarriers’ configuration of the the presence of user 2 and user 1 signals by considering them
proposed scheme is accomplished at the RX end where the as a noise, which degrades the BER performance.
wide subcarrier users (user 1 and user 3) are multiplexed by In the second scenario, user 2 and user 3 experience the
adopting an extended FFT operation. same channel effect. Therefore, if the fairness is a system
In Figure 1(d), user 3 symbol is multiplexed with the first requirement, both users have to be assigned similar PA.
half of user 2 symbol and user 1 does the same with the second However, the natural work of SIC depends on the power
half. This can be seen from Figure 1(b) as well, where user 2 differences to facilitate the separation process. Thus, with
has a contribution from user 1 and user 3 at the peaks. Note similar PA, the internal interference cannot be avoided and
that, user 1 and user 3 are orthogonal with respect to each the performance can be extremely degraded.
4 Wireless Communications and Mobile Computing

h1 h3
h1 h2
h2 h3

User 3
Base Station
Base Station User 1
User 1 User 2 User 3
User 2

(a) First scenario (b) Second scenario

Figure 2: Two different user distribution scenarios: (a) |ℎ1,𝑤 | > |ℎ2,𝑤 | > |ℎ3,𝑤 | and (b) |ℎ1,𝑤 | > (|ℎ2,𝑤 | = |ℎ3,𝑤 |).

3.2. Proposed NOMA (PN). According to the first scenario, where 𝑧𝑖,𝑤 represents the additive white Gaussian noise
by assigning user 2 narrow subcarriers, we reduce the (AWGN) for the 𝑖th user on subcarrier 𝑤 with a zero mean
2 2
interference imposed on user 1 and user 3 signals. Thus, SIC and 𝜎𝑖,𝑤 variance, that is, 𝑧𝑖,𝑤 ∼ N(0, 𝜎𝑖,𝑤 ), and ℎ𝑖,𝑤 denotes
process becomes easier since it has to differentiate two signals the channel gain between the BS and the received user at
rather than three based on their power differences. Actually, the 𝑤th subcarrier including both effects of large and small
the narrow subcarrier user is selected so that the other scale fading. Path loss and shadowing are considered as small
users, that is, wide subcarrier users, obtain more distinctive fading effects. On the other hand, block Rayleigh is adopted
channel conditions. Besides, the PA of user 2 is determined for large-scale fading.
independently which grants more flexibility for the system Without loss of generality, the channels are sorted as
design.
|ℎ1,𝑤 |2 > |ℎ2,𝑤 |2 > ⋅ ⋅ ⋅ > |ℎ𝑖,𝑤 |2 > ⋅ ⋅ ⋅ > |ℎ𝐼,𝑤 |2 > 0. For a given
This becomes very beneficial for the second scenario
subcarrier, a user who enjoys a better downlink channel
where achieving the fairness is an issue. As mentioned before,
quality can decode and remove the CCI from a user who has
user 2 and user 3 cannot be paired as they have similar
channel conditions. Nevertheless, if user 2 is assigned narrow a worse downlink channel quality by employing SIC [5]; thus,
subcarriers, user 3 and user 1 can be paired and a proper PA user 𝑖 enjoys a better channel quality than user (𝑖 + 1). At
is determined. Naturally, the PA of user 2 is not restricted by the 𝑖th user, if SIC is carried out perfectly, then achieving the
the other users. fairness follows Shannon’s equation [13], where the achievable
rate of the 𝑖th user for 𝐵tot Hz system bandwidth at the RX side
is given by
4. Conventional NOMA System Model
𝑁sc
Conventional multicarrier downlink NOMA system is for- 𝑅𝑖 = 𝐵sc ∑ log2 (1 + SINR𝑖,𝑤 ) ,
mulated by considering 𝐼 downlink users around a BS as 𝑤=1
shown in Figure 2. Users are distributed randomly and served (2)
󵄨 󵄨2
by one BS and the total bandwidth 𝐵tot consists of 𝑁sc number 𝑃𝑖,𝑤 󵄨󵄨󵄨ℎ𝑖,𝑤 󵄨󵄨󵄨
SINR𝑖,𝑤 = ( 󵄨 󵄨2 𝑖−1 ),
of orthogonal subcarriers in frequency domain. Transceivers 󵄨󵄨ℎ𝑖,𝑤 󵄨󵄨 ∑ 𝑃𝑢,𝑤 + 𝜎2
󵄨 󵄨 𝑢=1 𝑖,𝑤
are supposed to be equipped with one antenna, and 𝐼 users
share 𝑁sc OFDM subcarriers through superposition coding. where SINR𝑖,𝑤 is the instantaneous signal-to-interference-
The BS assigns different power levels depending on users’ plus-noise ratio by user 𝑖 on the 𝑤th subcarrier and 𝐵sc =
conditions in order to achieve the fairness in the system and 𝐵tot /𝑁sc is the subcarrier bandwidth. Note that if the strongest
provide the capability of detection at the RX side using SIC. user, that is, user 1, decodes and cancels all other users’ signals
So, high power is assigned to the poor users, that is, far users, successively, then the achievable data rate is given by 𝑅1 =
and low power to ones whose channel conditions are good. 𝐵sc ∑sc 2 2
𝑤=1 log2 (1 + 𝑃1,𝑤 |ℎ1,𝑤 | /𝜎1,𝑤 ).
In other words, NOMA exploits the heterogeneity of users’
distribution and then allows the separation in the power
domain [11].
5. Signal Representation and Design Analysis
The BS is transmitting the signal 𝑥𝑖,𝑤 to the 𝑖th user (𝑖 = This section considers the design analysis of our proposed
{1, 2, . . . , 𝐼}) on the 𝑤th subcarrier (𝑤 = {1, 2, . . . , 𝑁sc }) with scheme. The mathematical model of the proposed scheme
transmission power 𝑃𝑖,𝑤 , and, then, the received signal by user is established. The time domain signal in Figure 1(d) is
𝑖 on subcarrier 𝑤 is given by [12] as follows: formulated. Then, it is shown that, by adopting FFT operation
𝐼 with a length of two OFDM symbols at the RX side, the wide
𝑦𝑖,𝑤 = ℎ𝑖,𝑤 ∑ √𝑃𝑢,𝑤 𝑥𝑢,𝑤 + 𝑧𝑖,𝑤 , (1) subcarrier users are composed although they are orthogo-
𝑢=1 nally transmitted.
Wireless Communications and Mobile Computing 5

Generation of wide subcarrier signals can be done using together forming another block 𝑥𝑥nr which is already with
an inverse fast Fourier transform (IFFT) process with a length a length of 𝑀 samples. Finally, the resultant blocks are added
of 𝑁 samples. On the other hand, narrow subcarrier signals synchronously to produce one block 𝑠 with a length of 𝑀
can be also generated using IFFT process with a length of 𝑀 = samples for the transmission; this process can be expressed
𝑄𝑁 samples, where 𝑄 is the ratio between the two different as follows:
subcarrier spacings or the two different symbol lengths. 𝑠 = 𝑥𝑥𝑤 + 𝑥𝑥nr ,
According to Figure 1(b), user 1 and user 3 signals are
generated using IFFT with 𝑁 points, while user 2 signal is 𝐵 (6)
generated utilizing IFFT with 𝑀 = 2𝑁 points. It is worth 𝑥𝑥𝑤 = [𝑥𝑤1 , . . . , 𝑥𝑤𝐴 ]𝑀=𝐴𝑁 , 𝑥𝑥nr = ∑ 𝑥𝑛𝑟𝑏 .
𝑏=1
mentioning that, to avoid direct interference with wide sub-
carriers, we do not use all of the narrow subcarriers for trans- According to the proposed scheme in Figure 1, 𝐴 = 2, 𝐵 = 1,
mission. In our example, narrow-odd subcarriers are used for 𝑀 = 2𝑁, and 𝑄 = 2. So (6) can be written as follows:
user 2 data transmission while narrow-even ones are filled
with zeros using subcarrier mapping (SM). 𝑥𝑥𝑤 = [𝑥𝑤1 , 𝑥𝑤2 ]2𝑁 ,
(7)
An OFDM transmission symbol, of wide subcarrier user,
𝑥𝑥nr = [𝑥nr1 ]𝑀=2𝑁 .
is given by the 𝑁 point complex modulation sequence
By assuming 𝑚 = 𝑛 + 𝑞𝑁, (4) can be expressed as follows:
𝑥𝑤𝑎 (𝑛) = √𝑃𝛼𝑎 IFFT (𝑋𝑤𝑎 )
1 𝑁−1
𝑥nr (𝑛 + 𝑞𝑁) = ∑ √𝑃𝛽1 𝑋nr (𝑄𝑘 + 1)
1 𝑁−1 𝑀 𝑘=0
= √𝑃𝛼𝑎 ∑ 𝑋𝑤𝑎 (𝑘) ⋅ 𝑒𝑗2𝜋𝑛𝑘/𝑁, (3)
(8)
𝑁 𝑘=0 ⋅ 𝑒𝑗2𝜋(𝑛+𝑞𝑁)(𝑄𝑘+1)/𝑀,
for 𝑛 = 0, 1, . . . , 𝑁 − 1, 𝑎 = 0, 1, . . . , 𝐴, for 𝑛 = 0, 1, . . . , 𝑁 − 1, 𝑞 = 0, 1, . . . , 𝑄 − 1.
where 𝑋𝑤𝑎 (𝑘) is the complex modulated symbol of 𝑎th user Since 𝑄 = 2, the first half and second half of 𝑥nr signal can be
on 𝑘th subcarrier; that is, 𝐴 is the number of wide subcarrier represented by setting 𝑞 to 0 and 1, respectively:
users, 𝛼𝑎 is the assigned PA factor to the 𝑎th user, and
1 𝑁−1
∑𝐴𝑎=1 𝑃𝛼𝑎 is the amount of power that is specified for wide 𝑥nr (𝑛) = ∑ √𝑃𝛽1 𝑋nr (𝑄𝑘 + 1)
subcarrier users. 𝑀 𝑘=0
In a similar way, the OFDM transmission symbol, for
narrow subcarrier user, is given by the 𝑀 point complex ⋅ 𝑒𝑗2𝜋𝑛(𝑄𝑘+1)/𝑀, for 𝑞 = 0,
modulation sequence
1 𝑁−1 (9)
𝑥nr (𝑛 + 𝑁) = ∑ √𝑃𝛽1 𝑋nr (𝑄𝑘 + 1)
̂ 𝑀 𝑘=0
𝑥nr𝑏 (𝑚) = √𝑃𝛽𝑏 IFFT (𝑋𝑏)
𝑀−1 ⋅ 𝑒𝑗2𝜋(𝑛+𝑁)(𝑄𝑘+1)/𝑀, for 𝑞 = 1,
1 ̂𝑏 (𝑙) ⋅ 𝑒𝑗2𝜋𝑚𝑙/𝑀,
= √𝑃𝛽𝑏 ∑ 𝑋 (4)
𝑀 𝑙=0
𝑥nr (𝑛 + 𝑁) = −𝑥nr (𝑛) .

for 𝑚 = 0, 1, . . . , 𝑀 − 1, 𝑏 = 0, 1, . . . , 𝐵. Thus, the second half of the signal 𝑥nr is just a reversal copy of
the first half because of odd subcarriers usage. From (9), the
̂𝑏 (𝑙)
𝑋 transmitted signal 𝑠 in (6) can be expressed as follows:

𝑙−1 (5) {𝑥𝑤1 (𝑟) + 𝑥nr1 (𝑟) , 0 < 𝑟 < 𝑁 − 1

{𝑋 ( ) , 𝑙 = 𝑄𝑘 + 1, (𝑙 = 1, 3, . . . , 𝑀 − 1) 𝑠 (𝑟) = { (10)
= { nr𝑏 𝑄 𝑥𝑤2 (𝑟) − 𝑥nr1 (𝑟) , 𝑁 < 𝑟 < 2𝑁 − 1,
{
{0, o.w.,
where 𝑟 represents the composed signal sample index (𝑟 =
where 𝑋̂𝑏 (𝑙) is the complex modulated symbol of 𝑏th user on {0, 1, . . . , 𝑀 − 1}).
𝑙th subcarrier after SM; that is, 𝐵 is the number of narrow After composing the signals, to avoid intersymbol inter-
subcarrier users, 𝛽𝑏 is the assigned PA factor to the 𝑏th user, ference and enable frequency domain equalization (FDE) at
∑𝐵𝑏=1 𝑃𝛽𝑏 is the amount of power that is specified for narrow the RX, a copy from the resultant tail is appended as a CP
subcarrier users, 𝐼 = 𝐴 + 𝐵 is the total number of users, and where its duration has to be larger than the maximum excess
the maximum assigned power from the BS to all users is 𝑃 = delay of the channel.
∑𝐴 𝐵
𝑎=1 𝑃𝛼𝑎 +∑𝑏=1 𝑃𝛽𝑏 ; that is, 𝐵 = 0 represents the conventional
At the RX end and after removing the CP, FFT operation,
NOMA scheme case. with a length of 𝑀 = 2𝑁 points, is performed as follows:
The wide subcarrier signals (𝑥𝑤 ) are assigned different 𝑀−1
time slots to form one block 𝑥𝑥𝑤 with a length of 𝑀 samples. 𝑆 (V) = ∑ 𝑠 (𝑟) ⋅ 𝑒−𝑗2𝜋𝑟V/𝑀 for V = 0, 1, . . . , 𝑀 − 1, (11)
Thereafter, the narrow subcarrier signals (𝑥nr ) are added 𝑟=0
6 Wireless Communications and Mobile Computing

xw1
√P1
S
IFFT
Xw1 /
N TX
P RX
P xxw
√P2 xw2 / S

S S s S FFT
IFFT − Down
Detect Detect Detect
Xw2 / s + / 2N Sample Xw2 SIC Xw1
CP S[QK]
N P FDE QK Signal signal
P CP


X xnr1
√P1
Xnr1 S P 
x xw
IFFT
/ S·M /
2N xxnr
P S Detect Xnr1

(a) TX side (b) RX side

Figure 3: Transceiver design for proposed NOMA scheme adopting three downlink users.

where 𝑆 are the received complex symbols after FFT opera- Table 1: Simulation parameters.
tion. Afterwards, FDE takes the responsibility to get rid of
Parameter name Value
channel’s sparsity where single tap equalization is available.
To compute the output on the even and odd subcarriers, we Number of wide subcarriers (𝑁) 64
assume that V = 𝑄𝑘 + 𝑞, and, then, (11) can be represented as Number of narrow subcarriers (𝑀) 128
Modulation type QPSK
𝑁−1
Total system bandwidth 5 MHz
𝑆 (𝑄𝑘 + 𝑞) = ∑ (𝑥𝑤1 (𝑟) + 𝑥nr1 (𝑟)) ⋅ 𝑒−𝑗2𝜋𝑟(𝑄𝑘+𝑞)/𝑄𝑁
𝑟=0
Total power 10 dBm
The first scenario |ℎ𝑖,𝑤 |2 /𝜎𝑖,𝑤
2
20 dB, 17 dB, 0 dB for 𝑖 = 1, 2, 3
2𝑁−1
(12) The second scenario |ℎ𝑖,𝑤 |2 /𝜎𝑖,𝑤2
20 dB, 0 dB, 0 dB, for 𝑖 = 1, 2, 3
+ ∑ (𝑥𝑤2 (𝑟) − 𝑥nr1 (𝑟))
𝑟=𝑁

⋅ 𝑒−𝑗2𝜋𝑟(𝑄𝑘+𝑞)/𝑄𝑁. of interferer users. Moreover, the interference imposed on

narrow subcarrier user can be eliminated by detecting and
By setting 𝑞 = 0 and assuming 𝑧 = 𝑟 − 𝑁 for the second part canceling wide subcarrier users and utilizing the unused
of (12), then, the output on even subcarriers is proven to be narrow subcarriers. Furthermore, the narrow subcarrier user
can enjoy any power level. Thus, the BER performance
𝑆 (𝑄𝑘) = √𝑃𝛼1 𝑋𝑤1 (𝑘) + √𝑃𝛼2 𝑋𝑤2 (𝑘) . (13) is enhanced significantly. Note that the SINR values are
considered at the RX side after FFT process.
Explicitly, (13) proves that although 𝑋𝑤1 and 𝑋𝑤2 signals are The SINR𝑐 𝑖,𝑤 values utilizing conventional NOMA
constructed orthogonally at the TX, they are multiplexed by scheme for user 1, user 2, and user 3 on subcarrier 𝑤, with
utilizing larger FFT window at the RX side. In addition, (13) successful decoding and no error propagation assumption,
ensures the absence of narrow subcarriers contribution to are given by
wide subcarriers; thereafter, SIC separates the wide subcarrier
󵄨 󵄨2
signals based on power differences. 𝛼1 𝑃 󵄨󵄨󵄨ℎ1,𝑤 󵄨󵄨󵄨
User 1 and user 3 signals are constructed again; then, SINR𝑐 1,𝑤 = ( 2
),
𝜎1,𝑤
the reconstructed signal 𝑥 ̂𝑥𝑤 is subtracted from the received
signal 𝑠 with a view to detect user 2 signal. Transceiver block 󵄨 󵄨2
𝛼2 𝑃 󵄨󵄨󵄨ℎ2,𝑤 󵄨󵄨󵄨
diagram is given in Figure 3 for the proposed NOMA scheme. SINR𝑐 2,𝑤 = ( 󵄨 ), (14)
󵄨󵄨ℎ2,𝑤 󵄨󵄨󵄨2 𝛼1 𝑃 + 𝜎2
󵄨 󵄨 2,𝑤
󵄨 󵄨2
6. Performance Evaluation 𝛼3 𝑃 󵄨󵄨󵄨ℎ3,𝑤 󵄨󵄨󵄨
SINR𝑐 3,𝑤 = ( 󵄨 ),
󵄨󵄨ℎ3,𝑤 󵄨󵄨󵄨2 (𝛼1 + 𝛼2 ) 𝑃 + 𝜎2
In this section, we evaluate the performance of proposed 󵄨 󵄨 3,𝑤
NOMA scheme through simulation. System parameters are
presented in Table 1. while SINR𝑝 𝑖,𝑤 values utilizing proposed NOMA scheme are
expressed as follows:
6.1. BER. The BER performance is supposed to be enhanced 󵄨 󵄨2
by performing proposed NOMA due to many reasons. 𝑝 𝛼1 𝑃 󵄨󵄨󵄨ℎ1,𝑤 󵄨󵄨󵄨
SINR 1,𝑤 =( 2
),
Mainly, wide subcarrier users experience lower number 𝜎1,𝑤
Wireless Communications and Mobile Computing 7

󵄨 󵄨2
𝑝 𝛽1 𝑃 󵄨󵄨󵄨ℎ2,𝑤 󵄨󵄨󵄨 100
SINR 2,𝑤 =( 2
),
𝜎2,𝑤
10−1
󵄨 󵄨2
𝛼2 𝑃 󵄨󵄨󵄨ℎ3,𝑤 󵄨󵄨󵄨 10−2
SINR𝑝 3,𝑤 = ( 󵄨 ).
󵄨󵄨ℎ3,𝑤 󵄨󵄨󵄨2 𝛼1 𝑃 + 𝜎2

BER
󵄨 󵄨 3,𝑤
10−3
(15)
10−4
Using the same PA for both schemes, we can notice from (14)
and (15) that the first user experiences the same SINR values, 10−5
while a big enhancement, in SINR values, is noticeable for the 6 8 10 12 14 16 18 20 22
second and third user utilizing proposed NOMA. SNR (dB)

CN , U1 , 1 = .01 PN , U1 , 1 = .08
6.2. Fairness Factor (𝐹). To evaluate the fairness level for CN , U2 , 2 = .15 PN , U2 , 1 = .36
conventional and proposed NOMA we define the factor 𝐹 as CN , U3 , 3 = .84 PN , U3 , 2 = .56
in [14], where 𝐹 measures the equality of users’ rate 𝑅 for a
Figure 4: BER performance utilizing the optimum PA for conven-
given system and it is given by tional and proposed NOMA in the first scenario, 𝐹 = 0.7.
2
(∑𝐼𝑖=1 𝑅𝑖 )
𝐹= 2
. (16)
𝐼 ∑𝐼𝑖=1 (𝑅𝑖 ) 100
For instance, if all users get the same amount of 𝑅, then the
10−1
value 𝐹 will be close to 1.
The goal of PA mechanism is to maximize the sum 10−2
BER

capacity under a fairness constraint for NOMA systems. The

optimization problem is formulated as 10−3

𝐼 𝑁sc
10−4
max 𝐵sc ∑ ∑ log2 (1 + SINR𝑖,𝑤 )
𝛼𝑎 ,𝛽𝑏
𝑖=1 𝑤=1 10−5
6 8 10 12 14 16 18 20 22
𝐼 𝑁sc
SNR (dB)
s.t: ∑ ∑ 𝑃𝑖,𝑤 ≤ 𝑃 (17)
𝑖=1 𝑤=1 CN , U1 , 1 = .02 PN , U1 , 1 = .13
CN , U2 , 2 = .39 PN , U2 , 1 = .34
𝑃𝑖,𝑤 ≥ 0, ∀𝑖, ∀𝑤 CN , U3 , 3 = .59 PN , U3 , 2 = .53

𝐹 = 𝐹, Figure 5: BER performance utilizing the optimum PA for conven-

tional and proposed NOMA in the second scenario, 𝐹 = 0.7.
where 𝐹 is the target fairness index in the network. The PA
coefficients (𝛼𝑎 , 𝛽𝑏 ) are obtained through exhaustive search
using algorithm 1 in [15].
0.85
According to the first scenario, the optimal PA coefficients
0.8
utilizing conventional NOMA and proposed NOMA schemes
are equal to [𝛼1 𝛼2 𝛼3 ] = [0.01 0.15 0.84] and [𝛼1 𝛽2 𝛼2 ] = 0.75
Fairness Factor

[0.08 0.36 0.56], respectively. On the other hand, based 0.7

on the second scenario, the optimal PA coefficients utiliz- 0.65
ing conventional NOMA and proposed NOMA are found 0.6
to be [𝛼1 𝛼2 𝛼3 ] = [0.02 0.39 0.59] and [𝛼1 𝛽2 𝛼2 ] = 0.55
[0.13 0.34 0.53], respectively.
0.5
Based on the obtained optimal PA coefficients, the BER
0.45
performance is evaluated. The normalized channel gains 1 2 3 4 5 6 7 8 9 10
(|ℎ𝑖,𝑤 |2 /𝜎𝑖,𝑤
2
) are set as in Table 1 and the fairness index 𝐹 is SNR (dB)
assumed to be 0.7. The individual BER for the first and the
second scenarios are shown in Figures 4 and 5, respectively. ０． (1 = .13, 1 = .34, 2 = .53)
＃． (1 = .02, 2 = .39, 3 = .59)
Using the optimal PAs obtained for the second scenario, the ０． (1 = .02, 2 = .39, 3 = .59)
fairness level, of conventional and proposed NOMA schemes, ＃． (1 = .13, 1 = .34, 2 = .53)
is evaluated as depicted in Figure 6. The results show clear
dominance of proposed NOMA over conventional NOMA in Figure 6: Fairness level of conventional and proposed NOMA
terms of BER and fairness level. utilizing the optimum PA for the second scenario.
8 Wireless Communications and Mobile Computing

7 between OFDM symbols became unnecessary. Thus, our

6 proposed scheme is more spectrally efficient than OMA
5 schemes.
SE (bps/Hz)

4
Conflicts of Interest
3
The authors declare that there are no conflicts of interest
regarding the publication of this paper.
2

0 5 10 15 20 Acknowledgments
SNR (dB)
The authors thank Khawar Hussain, Research Assistant,
＃．, (U1 , 1 = .01) , (U2 , 2 = .15) , (U3 , 3 = .84) for assistance and comments that greatly improved the
０．, (U1 , 1 = .08) , (U2 , 1 = .36) , (U3 , 2 = .56) manuscript.
OMA, P1 = P2 = P3

Figure 7: Average spectral efficiency adopting conventional References

NOMA, proposed NOMA, and OMA using the optimum PA for the
first scenario. [1] U. L. Rohde, A. K. Poddar, I. Eisele, and E. Rubiola, “Next
generation 5G radio communication NW,” in Proceedings of
the 2017 Joint Conference of the European Frequency and Time
Forum and IEEE International Frequency Control Symposium
6.3. Spectral Efficiency (SE). The spectral efficiency 𝜂𝑖 = ((EFTF/IFC), pp. 113–116, Besançon, France, July 2017.
𝑅𝑖 /𝐵tot represents the amount of the carried data over the [2] J. G. Andrews, S. Buzzi, and W. Choi, “What will 5G be?” IEEE
available resources. In [16], it is proven that the conventional Journal on Selected Areas in Communications, vol. 32, no. 6, pp.
NOMA schemes can offer a better SE than OMA. In con- 1065–1082, 2014.
ventional NOMA, the SE is expected to improve dramatically [3] L. Dai, B. Wang, Y. Yuan, S. Han, C. I, and Z. Wang, “Non-
since three users utilize the available resources, that is, time orthogonal multiple access for 5G: solutions, challenges, oppor-
and frequency units. Unfortunately, this comes at a price of tunities, and future research trends,” IEEE Communications
unguaranteed QoS and loss of fairness. On the other hand, Magazine, vol. 53, no. 9, pp. 74–81, 2015.
although proposed NOMA offers less SE than conventional [4] Z. Ding, Z. Zhao, M. Peng, and H. V. Poor, “On the Spectral
NOMA due to additional time unit usage, that is, two time Efficiency and Security Enhancements of NOMA Assisted
slots, it still provides superior SE compared with OMA. In Multicast-Unicast Streaming,” IEEE Transactions on Communi-
particular, the superior SE in proposed NOMA arises from cations, vol. 65, no. 7, pp. 3151–3163, 2017.
the capability of the narrow subcarrier user to share the [5] Y. Sun, D. W. K. Ng, Z. Ding, and R. Schober, “Optimal joint
available resources in addition to CP reduction between power and subcarrier allocation for full-duplex multicarrier
OFDM symbols. non-orthogonal multiple access systems,” IEEE Transactions on
Communications, vol. 65, no. 3, pp. 1177–1191, 2017.
The first scenario in Figure 2(a) is considered in the
[6] Z. Ding, P. Fan, and H. V. Poor, “Impact of User Pairing on
evaluation of the average SE for OMA, conventional NOMA,
5G Nonorthogonal Multiple-Access Downlink Transmissions,”
and proposed NOMA. For OMA, the bandwidth 𝐵tot and the IEEE Transactions on Vehicular Technology, vol. 65, no. 8, pp.
total power 𝑃 are split equally between the assigned users. The 6010–6023, 2016.
SE performance is illustrated in Figure 7. [7] S. Timotheou and I. Krikidis, “Fairness for Non-Orthogonal
It is obvious from Figure 7 that the SE of proposed NOMA Multiple Access in 5G Systems,” IEEE Signal Processing Letters,
is decreased compared to conventional NOMA; however, it is vol. 22, no. 10, pp. 1647–1651, 2015.
still more spectrally efficient than OMA. [8] S. M. R. Islam, N. Avazov, O. A. Dobre, and K.-S. Kwak,
“Power-Domain Non-Orthogonal Multiple Access (NOMA) in
5G Systems: Potentials and Challenges,” IEEE Communications
7. Conclusion Surveys & Tutorials, vol. 19, no. 2, pp. 721–742, 2017.
In this work, some of the NOMA-OFDM system based [9] Z. DIng, X. Lei, G. K. Karagiannidis, R. Schober, J. Yuan, and
problems have been addressed by employing numerology V. K. Bhargava, “A Survey on Non-Orthogonal Multiple Access
for 5G Networks: Research Challenges and Future Trends,” IEEE
concept cleverly. More DOF are given to one of the composed
Journal on Selected Areas in Communications, vol. 35, no. 10, pp.
users by assigning narrower subcarriers. Based on this DOF, 2181–2195, 2017.
the constraints associated with the conventional NOMA [10] J.-Y. Chouinard, X. Wang, and Y. Wu, “MSE-OFDM: A new
schemes have been reduced and the BER performance has OFDM transmission technique with improved system perfor-
been improved too. Furthermore, the proposed method has mance,” in Proceedings of the 2005 IEEE International Confer-
proven its superiority in affording a fairer rate allocation to ence on Acoustics, Speech, and Signal Processing, ICASSP ’05, pp.
the users compared to conventional method. III865–III868, USA, March 2005.
A new methodology of multiplexing the users, in power [11] A. Benjebbour, Y. Saito, Y. Kishiyama, A. Li, A. Harada, and
domain, has been implemented by adopting a larger FFT T. Nakamura, “Concept and practical considerations of non-
window at the RX end. As a result, the guard durations orthogonal multiple access (NOMA) for future radio access,” in
Wireless Communications and Mobile Computing 9

Proceedings of the 2013 21st International Symposium on Intel-

ligent Signal Processing and Communication Systems, ISPACS
2013, pp. 770–774, Japan, November 2013.
[12] Z. Wei, D. W. K. Ng, J. Yuan, and H.-M. Wang, “Opti-
mal Resource Allocation for Power-Efficient MC-NOMA with
Imperfect Channel State Information,” IEEE Transactions on
Communications, vol. 65, no. 9, pp. 3944–3961, 2017.
[13] C. E. Shannon, “Communication in the Presence of Noise,”
Proceedings of the IRE, vol. 37, no. 1, pp. 10–21, 1949.
[14] R. K. Jain, D. M. W. Chiu, and W. R. Hawe, “A quantitative
measure of fairness and discrimination for resource allocation
in shared computer system,” vol. 38, pp. 20-21, Eastern Research
Laboratory, Digital Equipment Corporation, Hudson, Mass,
USA, 1984.
[15] T. Manglayev, R. C. Kizilirmak, and Y. H. Kho, “Optimum
power allocation for non-orthogonal multiple access (NOMA),”
in Proceedings of the 10th IEEE International Conference on
Application of Information and Communication Technologies,
AICT 2016, Azerbaijan, October 2016.
[16] P. Sedtheetorn and T. Chulajata, “Spectral efficiency evaluation
for non-orthogonal multiple access in Rayleigh fading,” in
Proceedings of the 18th International Conference on Advanced
Communications Technology, ICACT 2016, pp. 747–750, kor,
February 2016.
 International Journal of

Rotating Advances in
Machinery Multimedia

The Scientific
Engineering
Journal of
Journal of

Hindawi
World Journal
Hindawi Publishing Corporation Hindawi
Sensors
Hindawi Hindawi
[Link] Volume 2018 [Link]
[Link] Volume 2018
2013 [Link] Volume 2018 [Link] Volume 2018 [Link] Volume 2018

Journal of

Control Science
and Engineering

Advances in
Civil Engineering
Hindawi Hindawi
[Link] Volume 2018 [Link] Volume 2018

Submit your manuscripts at

[Link]

Journal of
Journal of Electrical and Computer
Robotics
Hindawi
Engineering
Hindawi
[Link] Volume 2018 [Link] Volume 2018

VLSI Design
Advances in
OptoElectronics
International Journal of

International Journal of
Modelling &
Simulation
Aerospace
Hindawi Volume 2018
Navigation and
Observation
Hindawi
[Link] Volume 2018
in Engineering
Hindawi
[Link] Volume 2018
Engineering
Hindawi
[Link] Volume 2018
Hindawi
[Link] [Link] Volume 2018

International Journal of
International Journal of Antennas and Active and Passive Advances in
Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration
Hindawi Hindawi Hindawi Hindawi Hindawi
[Link] Volume 2018 [Link] Volume 2018 [Link] Volume 2018 [Link] Volume 2018 [Link] Volume 2018