NON ORTHOGONAL MULTIPLE ACCESS

ILYAS DANISH (LB1902202)

DECEMBER 2019

This technical report provides a broad overview of the topic Non Orthog-
onal Multiple Access (NOMA) which is one of the latest developments in the
field of Digital Communications. This report provides an overview of fun-
damental aspects of NOMA with explanation of its operational procedure.
Furthremore, the important aspects of user scheduling and power control for
NOMA are treated in detail. For advancements in NOMA, incorporation of
Multi-Input Multi-Output (MIMO) with NOMA is discussed.

1 Introduction
As the name suggests, Non Orthogonal Multiple Access (NOMA) is a multiple-
access technique. Multiple-access is a primary form of multi-user communi-
cation where multiple users share a common channel to establish commu-
nications with a common station. Legacy multiple-access schemes involve
Time Division Multiple Access (TDMA), Frequency Division Multiple Ac-
cess (FDMA) and Code Division Multiple Access (CDMA) [1]. One impor-
tant aspect prevailing in all legacy multiple-access schemes is the fact that
the participating users are allocated channels which are orthogonal to each
other based on time, frequency or pesudo-code based segregation. NOMA

1
has unique property that here the signals for different users are not segre-
gated based on any of the previously mentioned domains, rather power level
is used to distinguish between signals for different users.
The ever increasing trend towards internet based multi-media services over
wireless network combined with the dawn of Internet of Things (IoT) concept
has brought unprecedented demands of high throughput low latency [2], [3].
Therefore 3GPP has put forth multifold spectrum efficiency target for Radio
Access of 5G [4]. Power domain NOMA promises significant improvement in
spectral efficiency as compared to legacy Orthogonal Multiple Access (OMA)
technologies by segregating multiple user’s signals in power domain and al-
lowing them to share the same time, frequency and code resource block.

2 NOMA Operation
NOMA transmission involves superposition coding where message signals in-
tended for multiple users are provided transmit power level inversely propor-
tional to corresponding user channel condition and added to form a composite
signal [5], [6]. At the receiver end, having higher power incoming signal, the
poor channel user is able to receive its message by treating the other message
signal as interference while the better channel user is able to decode its mes-
sage by utilizing Successive Interference Cancellation (SIC) to remove the
stronger message signal intended for poor channel user and then decoding
its own message [7], [2]. Thus NOMA utilizes the difference in channel qual-
ity conditions of the served users and enables them to pick their intended
message from composite signal based on signal power level difference. In the
case of NOMA downlink, the signal generated by the Base Station (BS) after
superposition coding (called composite signal) is given by:
N q
X
S= an P x n , (1)
n=1

2
where xn is the message signal intended for user n and an is the portion of
power level P allocated by the BS to the said signal. The resulting signal
received by a particular user n is given by [8]:
N q
X
yn = hn an P x n + n n , (2)
n=1

where nn is the Additive White Gaussian Noise (AWGN) component. The

operation of NOMA for the case of downlink established between single BS
and two users is shown in figure below. Note that the whole frequency band
is always available for use to both users since the power domain is utilized
for required segregation. The aspects which are dominant in operation and
performance of NOMA are user scheduling and power control.

Figure 1: NOMA operation and Power/Frequency resource usage [9]

3 NOMA Power Control

NOMA power control is the intelligence involved in deciding how the total
allowed transmit power should be divided among the participating users.
As shown in equation 1, the power level alloted to the individual users is
not constant for the population. It is a significant requirement of NOMA
communication that the power alloted should be controled in such a fashion
that users with strong channel be given lesser power as compared to that for

3
users with weak channel. This is done by ranking the users in descending
order of channel quality and then allocating the power levels in the reverse
fashion. The logic behind such power allocation comes from the fact that
strong channel users can easily decode and subtract the signals intended
for weak channel users and then extract their own signal. This leads to
reasoning that they may be provided lower transmit power. On the other
hand, the users with weak channel need to consider the signal for strong
channel user as interference therefore more transmit power should be given
to them. Objective function of NOMA power allocation is generally aimed for
ensuring user data-rate QoS constraints fulfillment ([17], [18], [19]) or aiming
towards achieving NOMA user data-rate always greater than OMA data-rate
[20].It was shown in [8] that if this condition of power control is not fulfilled
then users will always face outage.Therefore, to avoid high outage probability,
the power allocation algorithm must ensure that the better channel user has
adequate SNR to cancel the signal intended for poor channel user [8], [21].

4 NOMA User Scheduling

In order to reduce NOMA induced co-channel interference between users, all
NOMA users do not share the same resource but they are generally grouped
into clusters where only the user within a cluster share the same resource
block while the resources for different clusters are orthogonal. Efficient user
scheduling and power control among same cluster users can thus mitigate
this co-channel interference effect and bring significant spectral efficiency
for NOMA [10], [11]. Unlike OMA, scheduling of NOMA users does not
give priority to better channel users but it gives importance to distinction
between channel conditions of scheduled users. This is because if the channel
conditions of scheduled users are not different, the throughput performance
of NOMA (for both SISO and MIMO case) reduces to OMA level [12]. It

4
was showed in [11] through probabilistic analysis that scheduling of users
having disparate channel conditions will result in better individual and overall
rate. For both downlink and uplink case in cellular scenario, authors of [13]
provided insight into importance of best channel and worst channel users
while clustering so as to achieve maximum throughput improvement from
NOMA.

5 NOMA Performance
The achievable rate of NOMA for each user is based on the power allocated
and interferers involved. The achievable rate for the user with strongest
channel is given by:
 
Rn = log 1 + ρ|hn |2 an , (3)

while that for other users is given by:

ρ|hn |2 an
!
Rn = log 1 + . (4)
ρ|hn |2 N
P
i=n+1 ai + 1

Overall, the it can be stated that the achievable rate of NOMA approaches to
that of a multi-user system where all resources are allocated to best channel
user [7]. The ergodic sum rate performance of NOMA for cellular scenario
is shown in figure below. It depicts two cases for different BS-user distance
(RD ) and power attenuation coefficient (α). It shows that NOMA can achieve
upper bound of throughput which is better than that of OMA scheme. It
can also be shown that for NOMA case, user n will have diversity order of n,
as opposed to diversity order of only 1 achievable using conventional OMA
schemes [8].

5
 Figure 2: Ergodic Sum-rate performance of NOMA [8]

6 NOMA Benefits and Constraints

For NOMA, the resource sharing procedure better and poor channel users re-
sults in efficient utilization of resources since no resource is tied only to poor
channel users, on the other hand this feature also ensures fair and timely ser-
vice to all users [7]. Thus, owing to its enhanced spectrum efficiency along
with high connectivity gains and compatibility with legacy radio access tech-
nologies, NOMA is being considered as a suitable candidate for radio access
of 5G [12]. High throughput, low latency and more connectivity thus makes
NOMA a promising candidate for future mobile networks. On the other
hand, primary constraints of NOMA communication are high computational
complexity caused by SIC and relatively lower performance improvement de-
picted for low SNR range. Authors of [6] have provided elaborate overview
of various NOMA aspects including implementation, power allocation, user
pairing and performance analysis.

6
7 Advancements: MIMO-NOMA
Multi-Input Multi-Output NOMA (MIMO-NOMA) implementation based
on transmit beamforming has the combined advantage of capacity enhance-
ment from beamforming and spectral efficiency from NOMA. For each in-
dividual beam, It involves clustering (or scheduling) multiple users to be
simultaneously served using NOMA [14]. Performance analysis provided in
[10] shows that MIMO-NOMA is capable of marked gain over MIMO-OMA
in terms of ergodic sum capacity but proper power control and user admission
play a vital role in this performance improvement. System implementation
with performance analysis of transmit beamforming based MIMO-NOMA is
shown in [1]. Zero Forcing (ZF) beamforming provides a simple yet practical
option for MIMO-NOMA implementation and has been thoroughly investi-
gated in this regards [14], [15] and [10]. The study of MIMO-NOMA has
been extensively expanded regarding beamforming, power control and user
scheduling [16].

References
[1] John J. Proakis. Digital communications. 4th Edition.

[2] Y. Saito, Y. Kishiyama, A. Benjebbour, T. Nakamura, A. Li, and

K. Higuchi. Non-orthogonal multiple access (noma) for cellular future
radio access. In 2013 IEEE 77th Vehicular Technology Conference (VTC
Spring), pages 1–5, June 2013.

[3] S. M. R. Islam, N. Avazov, O. A. Dobre, and K. Kwak. Power-domain

non-orthogonal multiple access (noma) in 5g systems: Potentials and
challenges. IEEE Communications Surveys Tutorials, 19(2):721–742,
Secondquarter 2017.

7
 [4] J. G. Andrews, S. Buzzi, W. Choi, S. V. Hanly, A. Lozano, A. C. K.
Soong, and J. C. Zhang. What will 5g be? IEEE Journal on Selected
Areas in Communications, 32(6):1065–1082, June 2014.

[5] H. Zhang, F. Fang, J. Cheng, K. Long, W. Wang, and V. C. M. Le-

ung. Energy-efficient resource allocation in noma heterogeneous net-
works. IEEE Wireless Communications, 25(2):48–53, April 2018.

[6] L. Dai, B. Wang, Z. Ding, Z. Wang, S. Chen, and L. Hanzo. A survey of

non-orthogonal multiple access for 5g. IEEE Communications Surveys
Tutorials, 20(3):2294–2323, thirdquarter 2018.

[7] Z. Ding, Y. Liu, J. Choi, Q. Sun, M. Elkashlan, C. I, and H. V. Poor.

Application of non-orthogonal multiple access in lte and 5g networks.
IEEE Communications Magazine, 55(2):185–191, February 2017.

[8] Z. Ding, Z. Yang, P. Fan, and H. V. Poor. On the performance of non-

orthogonal multiple access in 5g systems with randomly deployed users.
IEEE Signal Processing Letters, 21(12):1501–1505, Dec 2014.

[9] Zhiguo Ding. Non-orthogonal multiple access (noma): Evolution to-

wards 5g cellular networks. Lancaster University, 2016.

[10] B. Kim, W. Chung, S. Lim, S. Suh, J. Kwun, S. Choi, and D. Hong. Up-
link noma with multi-antenna. In 2015 IEEE 81st Vehicular Technology
Conference (VTC Spring), pages 1–5, May 2015.

[11] Z. Ding, P. Fan, and H. V. Poor. Impact of user pairing on 5g nonorthog-

onal multiple-access downlink transmissions. IEEE Transactions on Ve-
hicular Technology, 65(8):6010–6023, Aug 2016.

[12] Z. Ding, X. Lei, G. K. Karagiannidis, R. Schober, J. Yuan, and V. K.

Bhargava. A survey on non-orthogonal multiple access for 5g networks:

8
 Research challenges and future trends. IEEE Journal on Selected Areas
in Communications, 35(10):2181–2195, Oct 2017.

[13] M. S. Ali, H. Tabassum, and E. Hossain. Dynamic user clustering and

power allocation for uplink and downlink non-orthogonal multiple access
(noma) systems. IEEE Access, 4:6325–6343, 2016.

[14] S. Ali, E. Hossain, and D. I. Kim. Non-orthogonal multiple access

(noma) for downlink multiuser mimo systems: User clustering, beam-
forming, and power allocation. IEEE Access, 5:565–577, 2017.

[15] B. Kimy, S. Lim, H. Kim, S. Suh, J. Kwun, S. Choi, C. Lee, S. Lee,

and D. Hong. Non-orthogonal multiple access in a downlink multiuser
beamforming system. In MILCOM 2013 - 2013 IEEE Military Commu-
nications Conference, pages 1278–1283, Nov 2013.

[16] Y. Huang, C. Zhang, J. Wang, Y. Jing, L. Yang, and X. You. Signal pro-
cessing for mimo-noma: Present and future challenges. IEEE Wireless
Communications, 25(2):32–38, April 2018.