Dynamic Power Allocation As A Way of Improving The Performance of Users in Non-Orthogonal Multiple Access (NOMA) For 5G Communications

Volume 8, Issue 5, May 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Dynamic Power Allocation as a Way of Improving

the Performance of Users in Non-Orthogonal Multiple
Access (NOMA) for 5G Communications
Mwewa Mabumba, Simon Tembo
The school of Engineering,
Department of Electrical & Electronics Engineering,
University of Zambia, Box 37392, Zambia
Abstract:-In this work the performance of the downlink realised via power domain, code domain and other
non-orthogonal multiple access (NOMA) technique is domains. NOMA offers a lot of advantages as compared to
investigated for two users, that is, cell centre user (user the conventional orthogonal multiple access (OMA) which
c) and cell edge user (user e). The performance of these has been used in the past generations, such as the fourth
users is highly dependent on the power split among the generation (4G). The conventional OMA technology is
data flows and the associated power allocation (PA) limited by orthogonal resources and it is difficult to meet
problem. In this research we propose a power allocation the need of the increasing user number. In NOMA systems,
scheme that ensures fairness for the downlink users Base Stations (BSs) transmit signals via superposition
ensuring improved quality of service (QoS). The system coding at the same time, frequencyand code but with
is analysed based on outage probability which is different power levels relying on successive interference
dependent on the channel state information (CSI) and is cancellation (SIC) performed at the receivers [3], [4]:
compared to the fixed power allocation scheme. The therefore NOMA provides higher spectral efficiency,
simulation results demonstrate the improvement in massive connectivity and low latency as compared to
performance of the users when dynamic allocation is OMA[5]. The users with good channel conditions are
used as compared to the fixed power allocation scheme. called strong users and the others are called weak users.
The weak users are allocated with more power whereas
Keywords:- NOMA, 5G, Power allocation, Outage strong users are allocated with less power for the purpose
probability of improving the user fairness. Weak users decode their
own messages by treating the strong users' messages as
I. INTRODUCTION noise. On the other hand, strong users implement SIC
Non-orthogonal multiple access (NOMA) scheme has technique to decode their own messages by removing the
received significant attention for fifth generation (5G) weak users' messages from the received signal[6]. User
cellular networks in the recent years [1], [2]. NOMA can fairness can be supported by appropriate allocation of
support multiple users within a single resource and thus can power coefficients at the base station. Fig. 1 shows a two
improve user and overall system throughput. It can be user NOMA downlinksystem.
Fig. 1: An example of two user NOMA downlink system
IJISRT23MAY829 www.ijisrt.com 928

ISSN No:-2456-2165
II. RELATED WORKS We assume the transmission power is P,𝑆𝑖 is the
message for the users and then the transmitted signal from
User fairness is a very important performance metric the base station can be written as:
of NOMA and has significantly received attention in the
recent years. NOMA raises a critical user fairness issue 𝑥 = ∑𝑛𝑖=1 √𝑃𝛼𝑖 𝑆𝑖 (1)
between cell centre and cell edge users. Fairness can be
supported through appropriate power allocation of the The received signal at user iis given by
superimposed transmitted data flows The user fairness
issues of NOMA are discussed in [4] with instantaneous or 𝑦𝑖 = ℎ𝑖 𝑥 + 𝑛𝑚 (2)
statistical CSI at the base stationand they formulated
algorithms that yield the optimal solution in both cases Where: ℎ𝑖 is the channel coefficient between BS and
considered. The power allocation for NOMA based on user i, 𝑛𝑚 represents the Gaussian noise plus interference.
proportional fairness scheduling is studied for both max-
Following the principal of NOMA the rates for user e
sum-rate and max-min-rate in [7]. A new definition of
and c in downlink NOMA are given by:
fairness which is purely based on information theoretic
grounds in the power domain NOMA wasstudied in[6]. A
𝛼𝑒 |ℎ𝑒|2
reversed relay assisted NOMA considering user fairness 𝑅𝑒 = log 2 (1 + 1 ) (3)
𝛼𝑒 |ℎ𝑒|2 +
was studied in[8]and the considered model analyzed in 𝜌
terms of all key performance indicators (KPIs) like ergodic

capacity, outage probability, and bit error rate with a more 𝑅𝑐 = log 2 (1 + 𝜌𝛼𝑐 |ℎ𝑐 |2) (4)
accurate imperfect SIC model and imperfect CSI. Overall respectively, where: 𝜌 = transmit SNR, 𝛼𝑒 &𝛼𝑐 =
the above studies show that NOMA can improve the user power allocation coefficients for user e & user c, and𝛼𝑒 >
fairness in terms of outage probability performance 𝛼𝑐 , 𝛼𝑒 + 𝛼𝑐 = 1.
compared to OMA, however they do not employ the
dynamic power allocation but instead use fixed power In this paper we assume user c with better channel
allocation which allocates power to the users utilizing a conditions as the strong user whereas user e with poor
fixed ratio based on their positions in the channel ordering. channel conditions can be regarded as a weak user, and
Thus the writers in [9] studied the performance assume that the target rate at user e is 𝑅𝑒𝑇 which means.
improvement for a cell – edge user in a two user noma
system by proposing two cooperativerelaying schemes but 𝑅𝑒 ≥ 𝑅𝑒𝑇 (5)
this introduces a performance loss for the cell center user.
Thus the proposed power allocation is derived by
The main novelty of this paper is to analyze NOMA working out the equations given.
system in terms of user fairness betweencell center and cell
edge users using proposed dynamic power allocation Note that this power allocation becomes a function of
scheme (proposed DPA) which is. Then based on this instantaneous channel gains meanwhile the fixed power
proposed power allocation scheme the system is further allocation factors are constant and not changing with
analyzed under the outage probability. channel gains.
III. SYSTEM MODEL
We will consider a two user downlink NOMA system

having a base station (BS), cell centre user (User C) and
cell edge user (User E). All wireless links assumed to
undergo Rayleigh fading with channel coefficients ordered
as in [10]|ℎ1 |2 ≤ |ℎ2 |2 ≤ ⋯ ≤ |ℎ𝑚 |2. If|ℎ1 |2 ≤ |ℎ2 |2 , then
more power is assigned to user e and less power is assigned
to user c. It is also assumed that the further away you move
from the base station the signal strength diminishes and
need for repeaters to boost the signal.
The composite signal having symbols of both users is
transmitted from the BS as seen in figure 2, which
represents the signal flow and process of obtaining the
optimal power in the downlink NOMA system.

ISSN No:-2456-2165
Power domain
Superposition
Xc Modulation
Xe Modulation
Channel
Rc Demodulation SIC Decoding
Re Demodulation Decoding of Xe
Subtracts user User c’s signal

e’s signal decoding
Fig. 2: Signal flow diagram of NOMA with SIC
A. Performance analysis – Outage probability (OP) and:

The system is analyzed based on outage probability in
order to evaluate its performance. The outage probability of 𝑃𝑒𝐷 = 1 − {𝑅𝑒,𝐷
𝑁
> 𝑅𝑒 } (7)
a communication channel can be defined as the probability
that the SNR of a channel falls below a predefined target 𝑁 𝛼𝑒 |ℎ𝑒|2
Where; 𝑅𝑒,𝐷 = log 2 (1 + 1 )
data rate[9]. Since NOMA allows users to share the same 𝛼𝑐 |ℎ𝑒 |2 +
𝜌
bandwidth and hence the data rate of each user must be
within the limits of channel capacity. The decoding rate of IV. SIMULATION RESULTS AND DISCUSSION
data by a user should be high so that latency caused by SIC
can be compensated. If the data rate of a user exceeds the In this section, we present the simulation results to
Shannon’s’ rate, outage occurs, which leads to loss of illustrate the achievable performance of the proposed
data[11]. The generalized expression for probability of scheme. In our plots, unless otherwise stated, the distance
outage is given below: between BS and both users is 10m and the path loss
exponent 𝜖 = 6. Fig. 3 and 4 we have results showing the
𝑃𝑐𝐷 = 1 − {𝑅𝑐→𝑒,𝐷
𝑁 𝑁
> 𝑅𝑒 , 𝑅𝑐,𝐷 > 𝑅𝑐 } (6) comparison of outage probability of a two user Noma
system. In figure 3, fixed power allocation is being used
Where; and𝛼𝑒 = 0.95, 𝛼𝑐 = 0.05. The wireless channel is
extremely dynamic in nature and fixed PA does not worry
𝑁 𝛼𝑒 𝜌|ℎ𝑐|2 𝑁
𝑅𝑐→𝑒,𝐷 = log 2 (1 + 𝛼 2
), 𝑅𝑐,𝐷 = log 2(1 + 𝜌𝛼𝑐 |ℎ𝑐 |2 ) about the instantaneous channel conditions of users.
𝑐 𝜌|ℎ𝑐| +1
Fig. 3: OPs of user c and e as a function of 𝑅∗ (bps) with the transmit SNR = 30dBm in FNOMA
Whenever the channel changes, the values of 𝛼𝑒 &𝛼𝑐 consideration the channel state information and the target
are fixed and as the target rate (𝑅∗ )exceeds two, the rate requirements into account. The probability of the
receiver is always in outage. Thus it can be seen from fig. 3 system failing is quite high all the time.
that the fixed power allocation termed FNOMA is
performing very poorly because it does not take into

ISSN No:-2456-2165
In fig. 4, the users are at the same distance of 10m using fixed power allocation. It is worth noting that there’s
from the BS. It can be seen that if we compare fig. 4 to fig. lower outage probability because 𝛼𝑒 &𝛼𝑐 aredynamically
3, there is an improvement in the performance of the adjusted based on target rate requirements and channel
system when the proposed dynamic power allocation state information. Whenever the channel changes, the
(proposed DPA) is used and the rate of failure for user c is values of 𝛼𝑒 &𝛼𝑐 are updated to meet the specifications
better as compared to that of user e, but both user e and reducing the rate of failure of the system.
user c in proposed DPA perform way better as compared to
Fig. 4: OPs of user c and e as a function of 𝑅∗ (bps) with the transmit SNR = 30dBm in DPA
V. CONCLUSION [5.] Y. Siato, Y. Kishiyama, A. Benjebbour, T.

Nakamura, A.Li, and K. Higuchi, "Non-orthogonal
In this research, we have studied performance multiple access (NOMA) for cellular future radio
improvement for users in a two user NOMA system. In Access," in 77th Vehicular Technology conference,
particular we have proposedDPA where the power 2013.
allocation coefficients are dynamically adjusted based on [6.] G. Gui, H. Sari, and E. Biglieri, "A New Definition
the state of the channel unlike fixed power allocation which of Fairness for Non-Orthogonal Multiple Access,"
always allocates more power to the cell edge user and the IEEE commun. Lett, vol. 23, no. 7, pp. 1558-2558,
system performance analyzed based on outage probability. 2019.
It is seen that the proposed DPA seems to be performing [7.] J. Choi, "power allocation for max-sum-rate and
better than the fixed power allocation in terms of outage. max-min-rate proportional fairness in NOMA,"
The rate of failure when fixed power allocation is used is IEEE Commun. Lett., vol. 20, no. 10, pp. 2055-2058,
very high as compared to proposed DPA. The results can 2016.
further be improved by employing other technologies like [8.] F. Kaya and H. Kaya, "Improved user Fairness in
D2D, MIMO, cooperative relaying etc in collaboration with decode - forward relaying non orthogonal multiple
NOMA. access schemes with imperfect SIC and CSI," IEEE
access, vol. 8, pp. 97542-97545, 2020.
REFERENCES [9.] T. N. Do, D. B. da costa, T.Q. Duong, and B. An,
"Improving the Performance of Cell - Edge Users in
[1.] S.M.R Islam, N. Avazov, O.A. Dobre, and K.S NOMA systems using Cooperative Relaying," IEEE
Kwak, "Power-domain non-orthogonal multiple J. Sel. Areas Communication, vol. 10, pp. 0090-
access (NOMA) in 5G Sytems: potencials and 6778, 2018.
challenges," Communication surveys, vol. 2, no. 99, [10.] J. Li and Q. Wang, "Power Allocation Method of
p. 1, 2016. Downlink Non-orthogonal Multiple Access System
[2.] 3GPP, "Initial views and evaluation results on non- Based on alpha Fair Utility Function," J Inf. Process
orthogonal multiple access for NR uplink," in R1- Syst., vol. 17, no. 2, pp. 306-309, 2021.
163111, 2016. [11.] L. Bhardwaj, R. K. Mishra, and R. Shankar , "Sum
[3.] Z. Ding, Y. Liu, J. Choi, Q. Sun, M. Elkashlan, I. rate capacity of non-orthogonal multiple access
Chih-Lin, and H. V. Poor, "Application of non- scheme with optimal power allocation," Defense
orthogonal multiple access in LTE and 5G Modelling and Simulation, 2021.
networks," IEEE Commun. Mag., vol. 55, no. 2, pp.
185-191, 2017.
[4.] S. Timotheou and I. Krikidis, "Fairness for non-
orthogonal multiple access in 5G systems," IEEE
Signal Process. Lett., vol. 22, no. 10, pp. 1647-1651,
2015.

Dynamic Power Allocation As A Way of Improving The Performance of Users in Non-Orthogonal Multiple Access (NOMA) For 5G Communications

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dynamic Power Allocation As A Way of Improving The Performance of Users in Non-Orthogonal Multiple Access (NOMA) For 5G Communications

Uploaded by

Copyright:

Available Formats

Volume 8, Issue 5, May 2023 International Journal of Innovative Science and Research Technology

Dynamic Power Allocation as a Way of Improving

Fig. 1: An example of two user NOMA downlink system

IJISRT23MAY829 www.ijisrt.com 928

terms of all key performance indicators (KPIs) like ergodic

III. SYSTEM MODEL

We will consider a two user downlink NOMA system

IJISRT23MAY829 www.ijisrt.com 929

Subtracts user User c’s signal

Fig. 2: Signal flow diagram of NOMA with SIC

A. Performance analysis – Outage probability (OP) and:

IJISRT23MAY829 www.ijisrt.com 930

V. CONCLUSION [5.] Y. Siato, Y. Kishiyama, A. Benjebbour, T.

IJISRT23MAY829 www.ijisrt.com 931

You might also like