Review Intelligent Reflective Surface Based Non Orthogonal Multiple Access (1).pdf

A Review: Intelligent Reflective Surface Based Non orthogonal Multiple Access
A REVIEW: INTELLIGENT
REFLECTIVE SURFACE BASED NON
ORTHOGONAL MULTIPLE ACCESS
Ms Manaswi M Latthe Dr Taherbasha Shaikh
Assistant Professor Professor
SITOE, Yadrav SR University, Warangal
India India
Abstract
Non-orthogonal multiple access (NOMA) plays a crucial role in 5th generation (5G) technology. A key advantage of
NOMA is its ability to serve multiple users simultaneously using the same time, frequency, or code, but with
varying power levels. This approach results in a considerable increase in spectral efficiency compared to traditional
orthogonal multiple access methods. NOMA functions in conjunction with multiple-input multiple-output (MIMO)
technologies. As a cutting-edge multiple access (MA) technique, NOMA shows promise in enhancing the spectral
efficiency of mobile communication networks [1-4]. Unlike previous generations of mobile networks that utilized
time, frequency, or code domains for multiple access, NOMA employs the power domain. A significant drawback of
the traditional orthogonal multiple access (OMA) approach is its reduced spectral efficiency when bandwidth
resources, such as subcarrier channels, are assigned to users experiencing poor channel conditions. In contrast,
NOMA allows each user to access all subcarrier channels, enabling users with strong channel conditions to utilize
bandwidth resources allocated to those with poor channel conditions. This approach substantially enhances spectral
efficiency.
Introduction
NOMA enables the timely service of users with varying channel conditions, potentially meeting 5G's stringent
requirements for ultra-low latency and ultra-high connectivity. The integration of NOMA with multiple-input
multiple-output (MIMO) technologies offers diverse MIMO-NOMA designs, balancing reception reliability and
data rates. Cooperative NOMA leverages user cooperation, as some users in NOMA systems can act as relays,
knowing information sent to others. This approach can capitalize on the heterogeneous nature of future mobile
networks, where certain users may possess superior capabilities, such as more antennas. Additionally, the interaction
between NOMA and cognitive radio (CR) technologies is considered crucial for next-generation networks.
In contrast to conventional methods, NOMA allocates more transmission power to users with poor channel
conditions. Specifically, the message for the user with the weaker channel is given higher transmission power,
ensuring direct detection of their message by treating other users' information as noise. In recent years, interference
has emerged as a significant obstacle to the successful implementation of various end-user applications in fifth-
generation (5G) wireless networks. Despite the development and adoption of numerous communication protocols

and standards during this period, interference continues to hinder the provision of quality of service (QoS) to end-
users across different 5G applications. To address these challenges, this paper presents a comprehensive survey of
cutting-edge non-orthogonal multiple access (NOMA) variants that utilize power and code domains as the
foundation for interference mitigation, resource allocation, and QoS management in the 5G landscape. These
variants support future smart communication and are complemented by device-to-device (D2D), cooperative
communication (CC), multiple-input and multiple-output (MIMO), and heterogeneous networks (HetNets). Existing
literature suggests that NOMA can resolve many issues in current proposals, enabling contention-based grant-free
transmissions between various devices. The paper also provides an in-depth discussion of the key differences
between orthogonal multiple access (OMA) and NOMA in 5G. Furthermore, it analyzes several open issues and
research challenges associated with NOMA-based applications. The study concludes with a comparative analysis of
different existing proposals, offering readers valuable insights into the subject matter.
The rapid expansion and widespread adoption of intelligent devices, including smart phones, wearable technology,
sensors, and actuators in Internet-based applications, is generating an enormous volume of data traffic. These
applications, such as ultra-high-definition video transmission, live streaming, augmented and virtual reality, real-
time video calls, live conferences, and social media platforms, are placing a significant burden on the long-term
evolution (LTE) and long-term evolution-advanced (LTE-A) spectrum. This overload is causing a decline in the
spectral efficiency (SE) of the underlying network infrastructure. According to Cisco's predictions, by 2023,
approximately 14.7 billion devices will be connected to the Internet of Things (IoT) [1]. Furthermore, the compound
annual growth report [1] from 2018 forecasts that over 78 billion connected devices will be utilizing cellular
network services by the end of 2023, as illustrated in Fig. 1. To address this challenge, various research
organizations are actively exploring the potential of 5G wireless networks, which are anticipated to be fully
commercialized by 2023.
In an effort to enhance the quality of service (QoS) and quality of experience (QoE) for end-users, numerous
organizations and countries are investing considerable resources int Various collaborative initiatives, including IMT-
2020 (3GPP), 5GPPP/METIS (European Union), 5G Forums (Korea), and ARIB (Japan), have been established. In
2015, the International Telecommunication Union-Radiocommunication (ITU-R) formally designated the 5G
wireless network as IMT-2020 and outlined its initial concept, key characteristics, and applications [2]–[4]. The
initial phase of the 5G standard was examined under 3GPP Release 15, featuring the following key attributes [3]: i)
connection density of 10,00000/km2, ii) bandwidth of 1 to 2GHz, iii) data rate of 20/10Gbps (downlink/uplink), iv)
latency under 1ms, v) spectral efficiency of 120 bps/Hz, vi) throughput of 10Gbps, vii) utilization of Massive
MIMO to improve coverage and capacity, viii) real-time information processing and transmission at speeds up to 1-
2Gbps, ix) frequency band (mmWave) ranging from 30 to 300 GHz, x) mobility of 500Kmph, xi) area traffic
capacity of 10 Mbits/s/m2, and xii) energy efficiency (EE) 50 to 100 times greater compared to IMT-A.
ITU-R categorizes 5G network applications into three main groups: i) enhanced mobile broadband (eMBB)
connection, ii) massive machine-type communications (mMTC), and iii) ultra-reliable low latency communication
(URLLC). To fulfill 5G requirements, the integration of technologies such as software-defined network (SDN),
coordinated multipoint (CoMP), cooperative com The integration of various technologies such as D2D, VLC, M2M,
CRN, small cells, mMIMO, mmWave, and V2X is essential [3]–[5]. This combination allows multiple users to share
the same resource block, enhancing spectral efficiency. However, managing asynchronous machine-generated data
to provide extensive connectivity and varied QoS presents challenges. To address this, efficient MA techniques like
OMA and NOMA are necessary. OMA falls short due to bandwidth inefficiency and high signaling overhead.

Consequently, NOMA techniques are being utilized under 3GPP NR Rel.-14 (downlink) and 3GPP NR Rel.-15
(uplink) standards to tackle these issues in 5G environments.
NOMA allows multiple users to share a single resource block, and users can utilize multiple resource blocks to
improve data rates. At the transmitter, signals are multiplexed using superposition coding, while demultiplexing
occurs at the receiver through successive interference cancellation. SC improves sum rate, user fairness, and
scheduling flexibility, while SIC eliminates inter-user interference. NOMA is categorized into PD-NOMA and CD-
NOMA. PD-NOMA serves multiple users from a common orthogonal resource block based on channel conditions,
while CD-NOMA employs code spreading sequences for user differentiation. This paper examines and compares
various NOMA variants within these categories using different evaluation parameters in Sections 3 and 4.
Previous surveys [6]–[17] have typically focused on either CD-NOMA or PD-NOMA individually. To our
knowledge, no comprehensive survey has considered both categories together. This survey covers most NOMA
variants under PD-NOMA and CD-NOMA for 5G networks. A summary of existing survey articles follows.
Yunzheng et al. [6] examined NOMA and waveform modulation techniques for 5G, but failed to address NOMA's
compatibility with 5G. Islam et al. [7] investigated PD-NOMA's challenges and implementation issues, while Liu et
al. [8] focused on theoretical aspects, with limited exploration of CD-NOMA. Ding et al. [9], [10] explored NOMA's
applications, research, and future challenges in 5G. Dai et al. [11] evaluated various NOMA challenges and
solutions, but lacked a comprehensive NOMA classification. Basharat et al. [12] concentrated on NOMA's MA
schemes for 5G and various decoding methods. Aldababsa et al. [13] investigated uplink and downlink NOMA
transmission models, including MIMO and cooperative communication extensions, as well as design challenges and
approaches. However, they did not explore NOMA's compatibility with 5G techniques. Dai et al. [14] focused on
PD-NOMA, CD-NOMA, and all 15 NOMA schemes under Rel.-14 3GPP NR, examining NOMA's performance
gains in various scenarios. Nevertheless, they did not address D2D communication's physical design issues. Cai et
al. [15] studied OMA modulation techniques and NOMA MA schemes, comparing different NOMA schemes based
on performance metrics. However, they overlooked other challenging NOMA applications such as D2D
communication, M2M communication, and HetNets. Wang et al. [16] emphasized design issues of NOMA
techniques for user separation, focusing on latency and throughput in PD-NOMA and CD-NOMA, while also
elaborating on spatial, hybrid, and network domains. Wu et al. [17] discussed CD-NOMA's technological
requirements and explored its test cases in 5G networks, but did not examine PD-NOMA and its compatibility with
various 5G techniques. Table 3 summarizes these survey articles and highlights key differences from the proposed
survey. Fig. 2, Table 3, and Table 4 illustrate related survey articles and research gaps compared to the proposed
survey. Existing NOMA surveys have primarily focused on power and code domains. This survey comprehensively
examines different PD-NOMA and CD-NOMA variants proposed by various authors in the context of 5G
techniques applicable to diverse applications [18]–[22].
Reconfigurable intelligent surfaces (RISs), also referred to as intelligent reflecting surfaces (IRSs)
1
, have garnered considerable interest due to their ability to improve wireless network capacity and coverage by
intelligently modifying the wireless propagation environment. As a result, RISs are viewed as a key technology for
sixth-generation (6G) communication networks. This paper presents a thorough review of the current state of RIS
research, emphasizing their operational principles, performance assessment, beamforming design and resource
management, machine learning applications in RIS-enhanced wireless networks, and the combination of RISs with

other emerging technologies. We elucidate the fundamental principles of RISs from both physics and
communications standpoints, which forms the basis for our discussion on performance evaluation of multi-antenna
assisted RIS systems. Moreover, we methodically examine existing designs for RIS-enhanced wireless networks,
encompassing performance analysis, information theory, and performance optimization aspects. Additionally, we
review current research efforts that employ machine learning to address challenges in dynamic scenarios, such as
unpredictable wireless channel fluctuations and user mobility in RIS-enhanced wireless networks. Lastly, we
highlight major issues and research opportunities related to integrating RISs with other emerging technologies for
application in next-generation networks.
The growing demand for high-quality, widespread wireless services presents significant challenges to current
cellular networks. The fifth-generation (5G) communication systems are designed to address applications such as
rate-focused enhanced mobile broadband (eMBB), ultra-reliable, low latency communications (URLLC), and
massive machine-type communications (mMTC) services. The objectives for sixth-generation (6G) wireless
communication systems are anticipated to be groundbreaking, encompassing data-driven, immediate, ultra-massive,
and omnipresent wireless connectivity, as well as connected intelligence [1], [2]. To support these new applications
and services, innovative transmission technologies are essential.
Reconfigurable intelligent surfaces (RISs), also known as intelligent reflecting surfaces (IRSs) [3], [4] or large
intelligent surfaces (LISs) [5], [6], consist of an array of reflecting elements that reconfigure incoming signals. Due
to their ability to actively modify the wireless communication environment, RISs have become a central focus in
wireless communications research for addressing various challenges in different wireless networks [7], [8]. The
benefits of RISs include:
• Simple installation: RISs are passive devices made of electromagnetic (EM) material. As shown in Fig. 1, they can
be installed on various structures, including building facades, indoor walls [9], aerial platforms, roadside billboards,
highway poles, vehicle windows, and even pedestrians' clothing due to their cost-effectiveness.
• Improved spectrum efficiency: RISs can reconfigure the wireless propagation environment by offsetting power
loss over extended distances. They can create virtual line-of-sight (LoS) connections between base stations (BSs)
and mobile users by passively reflecting received signals. The enhancement in throughput is particularly notable
when the LoS link between BSs and users is likely to be obstructed by tall buildings. Through strategic deployment
and design of RISs, a software-defined wireless environment can be established.
Eco-friendly:
Unlike traditional relaying systems such as amplify-and-forward (AF) and decode-and-forward (DF) [10], an RIS
can modify incoming signals by adjusting the phase shift of each reflecting element, eliminating the need for a
power amplifier [11], [12]. This makes RIS deployment more energy-efficient and environmentally friendly
compared to conventional AF and DF systems.
a pressing need to classify existing research, which is a primary objective of this study. Versatility:

RISs are capable of supporting full-duplex (FD) and full-band transmission because they simply reflect
electromagnetic waves. Moreover, wireless networks enhanced with RIS technology are compatible with existing
wireless network hardware and standards. Given their appealing features, RISs are considered an effective approach
to address various challenges in commercial and civilian applications. Numerous recent studies have explored RISs
and their contributions, focusing on multiple application scenarios under diverse assumptions. Consequently, the
system models proposed by these investigations Research contributions often vary, creating
Figure 1 showcases the implementation of RISs in various wireless communication systems. Figure 1(a) depicts
RIS-enhanced cellular networks, where an RIS is utilized to circumvent obstacles between base stations and users.
This approach enhances service quality in heterogeneous networks and improves latency performance in mobile
edge computing (MEC) systems [14], [15].
Additionally, an RIS can function as a signal reflection hub in device-to-device (D2D) communication networks,
supporting extensive connectivity through interference reduction [16]. Alternatively, in physical layer security (PLS)
networks, an RIS can eliminate unwanted signals by intelligently designing passive beamforming techniques.
Furthermore, an RIS can be utilized to boost the signal strength for users at the cell edge and reduce interference
from neighboring cells [18]. It can also compensate for significant power loss over long distances in SWIPT
networks [19]–[22]. Figure 1 (b) showcases RIS-assisted indoor communications, where an RIS mounted on a wall
can enhance service quality in high-bandwidth scenarios like virtual reality. Moreover, a virtual RIS-aided LoS
connection can be established to ensure complete coverage in scenarios sensitive to obstructions, such as visible
light communications [23] and WiFi networks. Figure 1 (c) depicts RIS-enhanced unmanned systems. An RIS can
be employed to improve the performance of UAV-enabled wireless networks [24], cellular-connected UAV
networks [25], autonomous vehicular networks, and AUV networks by fully leveraging the aforementioned RIS
advantages. Figure 1 (d) illustrates RIS-enhanced IoT networks, where an RIS is used to support intelligent wireless
sensor networks [26], smart agriculture, and smart manufacturing.
Unlike previous studies, this paper offers a comprehensive examination of the principles behind RIS-enhanced
wireless networks and explores research opportunities for utilizing RISs in various applications, including
unmanned systems, non-orthogonal multiple access (NOMA) in RIS-enhanced wireless networks, and machine
learning in RIS-enhanced wireless networks. Our key contributions are as follows:
•
We present the core principles governing RIS operation and their interaction with electromagnetic (EM) signals. We
also review typical RIS functions and their associated principles, with a focus on patch-array based implementation
and a comparison between ray-optics and wave-optics perspectives.
•
We create performance evaluation methods for multi-antenna assisted RIS systems. We also summarize research
contributions by highlighting their benefits and constraints.
•
We examine RISs from an information theory standpoint, using this as a basis to review protocols and approaches
for the joint design of beam forming and resource allocation with various optimization goals. Furthermore, we
discuss major unresolved research questions. RIS Classifications

RISs can be implemented using meta material or patch-array technologies, with meta material-based RISs known as
meta surfaces. These can be positioned as reflecting/refracting RIS between the BS and user, or as waveguide RIS at
the BS. RISs can be reconfigured through electrical, mechanical, or thermal means. Based on energy consumption,
RISs are classified as passive-lossy, passive-lossless, or active. The active or passive nature of RISs influences their
performance capabilities. It's noteworthy that RISs cannot be entirely passive due to their inherent configurability.
We examine three key RIS operational modes: waveguide [27], refraction [28], and reflection [29]. Using Love's
field equivalence principle [30], reflected and refracted electromagnetic fields can be analyzed by introducing
equivalent surface electric and magnetic currents [28]. In these modes, the RIS transforms and emits waves (either
from an incident wave or waveguide feed) into a desired free-space propagating wave. The surface equivalence
principle (SEP), encompassing Love's field equivalence principle and Huygens' principle, will be discussed in
Section II-B.
Fig. 1 depicts a multi-user downlink scenario featuring two user categories: near users and cell-edge users. The latter
are presumed to lack a direct connection to the base station. The system employs SDMA, wherein the base station,
equipped with M antennas, creates K beam forming vectors (wk, 1 ≤ k ≤ K) to serve K near users (M ≥ K) using
zero forcing beam forming. To highlight the advantages of IRS-NOMA, it is further assumed that the K near users
are selected due to their channel vectors being mutually orthogonal, implying that the beam forming vectors (wk)
are ortho normal. Once wk is established, IRS-NOMA is employed to accommodate more users on these predefined
beams compared to traditional SDMA. For demonstration purposes, we consider that each beam, wk, supports an
additional user, referred to as user k', utilizing an IRS equipped with N reflecting elements, as illustrated in Fig. 1.
Furthermore, we assume that only user k' can receive signals from IRS k, as these IRS's are positioned near cell-
edge users.
The base station transmits K k=1 wk(α1sk+α2sk ), where sk represents the signal intended for user k , sk is the
signal meant for user k, αi signifies the power allocation coefficient, and α2 1 +α2 2 = 1. Given the assumption that
no direct connection exists between the IRS's and the near users, the performance analysis for user k mirrors that of
conventional NOMA systems. Consequently, this letter concentrates solely on the performance of user k. The signal
received by user k can be expressed as yk = hH kΘkGk K k=1 wk(α1sk + α2sk ) + wk , (1) where Gk represents the
N × M complex Gaussian channel matrix from the base station to the IRS linked to user k , hk denotes the complex
Gaussian channel vector from the IRS to user k , and wk signifies the noise. Θk is a diagonal matrix with its main
diagonal elements represented by βk,ie−jθk,i , where θk,i indicates the reflection phase shift and βk,i represents the
amplitude reflection coefficient [6], [7].
As with conventional NOMA, it is assumed that α1 ≤ α2. The signal-interference-plus-noise ratio (SINR) for user k
to decode its message is thus given by
SINRk = |θH k Dk hk| 2α2 2 |θH k Dk hk|2α2 1+K i=1,i=k |θH k Dk hi|2+ 1 ρ , (2) where ρ denotes the transmit
signal-to-noise ratio (SNR), hi = Gkwi, θk is an N × 1 vector containing the elements on the main diagonal of ΘH k ,
and Dk is a diagonal matrix with its diagonal elements derived from hH k . It should be noted that the SINR in (2)
implies that a cell-edge user has full knowledge of hk , Gk, and wk. The channel state information pertaining to IRS,
Gk and hk , is presumed to be available through the channel estimation methods described in [12]–[14]. Details
about the predetermined beamforming vectors, wk, can be conveyed to the user by its IRS through a dependable
control channel. Furthermore, we note that the SINR expression is more complex than that of traditional NOMA,
owing to the presence of the product of the complex Gaussian distributed random variables.
1) Waveguide RIS:

R. Smith et al. [27] conducted a theoretical study on waveguide-fed metasurfaces. The metasurface elements are
modeled as uncoupled magnetic dipoles, with each dipole's magnitude proportional to the product of the reference
wave and the element's polarizability. Beamforming is achieved by adjusting the polarizability. Each metasurface
element functions as a micro-antenna. Compared to traditional antenna arrays, the compact waveguide metasurface
is space-efficient and capable of transmitting wide-angle radiation patterns.
2) Refracting RIS:
Viktar S et al. [31] proposed a theoretical design for a perfectly refracting and reflecting metasurface. They
employed an equivalent impedance matrix model to optimize tangential field components on both sides of the
metasurface. Additionally, they explored three potential device implementations: self-oscillating teleportation
metasurfaces, non-local metasurfaces, and metasurfaces composed of lossless components exclusively. The authors
also addressed the critical role of omega-type bianisotropy in designing lossless-component realizations of perfectly
refractive surfaces.
3) RIS Reflection:
Dai et al. [29] created an innovative reflective metasurface with digital coding capabilities. The metasurface's
components incorporate varactor diodes that feature adjustable biasing voltages. By establishing multiple

predetermined digitized biasing voltage levels, each component can execute discrete phase shifts, enabling
beamforming for the reflected wave.
The subsequent portion of this section will concentrate on the functional principles of RISs that serve as reflectors.
CONCLUSION
This paper has examined recent studies on RIS-enhanced wireless networks for next-generation systems, focusing
on RIS operating principles, performance assessment of multi-antenna RIS systems, RIS beamforming and resource
allocation, machine learning applications in RIS-enhanced networks, and their integration with other key 6G
technologies. The benefits and limitations of RISs for communication purposes have been outlined. Additional
research is necessary to connect the intricate physical models of various RIS implementations with commonly used
communication models. We have analyzed the performance of multi-antenna assisted RIS systems by methodically
reviewing existing designs for RIS-enhanced wireless networks from performance analysis, information theory, and
optimization perspectives. Most current research assumes perfect CSI at the BS, RIS controller, and user levels.
However, obtaining CSI in RIS-enhanced wireless networks is a significant challenge, requiring substantial training
overhead. We have also explored existing research that applies machine learning to address the dynamic nature of
wireless environments, including random channel fluctuations and user mobility. Guidelines for developing ML-
empowered RIS-enhanced wireless networks have been discussed. The advantages of combining RISs with NOMA,
UAV-terrestrial networks, PLS, and SWIPT have been examined. RIS-enhanced wireless network research is still in
its early stages, offering numerous opportunities for significant contributions and advancements in this field.
References:
[1] Y. Saito, Y. Kishiyama, A. Benjebbour, T. Nakamura, A. Li, and K. Higuchi, “Non-orthogonal multiple access (NOMA)
for cellular future radio access,” in Proc. IEEE Vehicular Technology Conference, Dresden, Germany, Jun. 2013.
[2] J. Choi, “On multiple access using H-ARQ with SIC techniques for wireless ad hoc networks,” Wireless Personal Commun.,
vol. 69, pp. 187–212, 2013.
[3] J. Choi, “Non-orthogonal multiple access in downlink coordinated two-point systems,” IEEE Commun. Letters, vol. 18, no.
2, pp. 313–316, Feb. 2014.
[4] 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 Process. Letters, vol. 21, no. 12, pp. 1501–1505, Dec. 2014.
[5] 3GPP TD RP-150496: “Study on Downlink Multiuser Superposition Transmission”.
[6] White Paper, “Rethink Mobile Communications for 2020+,” FuTURE Mobile Communication Forum 5G SIG, Nov. 2014.
http://www.future-forum.org/dl/141106/whitepaper.zip.
[7] L. Dai and B. Wang and Y. Yuan and S. Han and C. I and Z. Wang, “Non-orthogonal multiple access for 5G: solutions,
challenges, opportunities, and future research trends,” IEEE Commun. Magazine, vol. 53, no. 9, pp. 74–81, Sept. 2015.
[8] Z. Ding, F. Adachi, and H. V. Poor, “The application of MIMO to non-orthogonal multiple access,” IEEE Trans. Wireless
Commun., vol. 15, no. 1, pp. 537-552, Jan. 2016.
[9] J. Choi, “Minimum power multicast beamforming with superposition coding for multiresolution broadcast and application
to NOMA systems,” IEEE Trans. Commun., vol. 63, no.3, pp. 791-800, Mar. 2015.

Review Intelligent Reflective Surface Based Non Orthogonal Multiple Access (1).pdf

More Related Content

Similar to Review Intelligent Reflective Surface Based Non Orthogonal Multiple Access (1).pdf (20)

Recently uploaded (20)

Review Intelligent Reflective Surface Based Non Orthogonal Multiple Access (1).pdf