PaperlayerView exact revision on arXiv

Source-linked AI summary

Interplay Between NOMA and Other Emerging Technologies: A Survey

Mojtaba Vaezi, Gayan Amarasuriya, Yuanwei Liu, Ahmed Arafa, Fang Fang, Zhiguo Ding

arXiv:1903.10489v2cs.IT

TL;DR

Next-generation networks need greater connectivity and diverse performance, while individual technologies face limitations in practical deployments. This survey synthesizes NOMA's integration with emerging wireless technologies, emphasizing mutual benefits, challenges, and future directions. It concludes that these combinations can address limits that the technologies cannot overcome singlehandedly.

  • Problem

    5G and beyond require massive connectivity and diverse throughput and latency support, while emerging wireless technologies have limitations that motivate their combination with NOMA.

  • Method

    The paper surveys research combining NOMA with massive MIMO, millimeter wave, cognitive and cooperative communications, security, energy harvesting, mobile edge computing, and machine learning.

  • Results

    The survey explains how combining NOMA with these technologies can overcome limits that the individual technologies cannot overcome alone.

  • Takeaways & Limitations

    The paper identifies challenges and future research directions for NOMA-based combinations with emerging technologies in 5G networks and beyond.

  • Takeaways & Limitations

    Practical NOMA performance depends on accurate CSI, while residual pilot-contamination interference can remain detrimental even in asymptotic massive-MIMO regimes.

Abstract

from arXiv · show

Non-orthogonal multiple access (NOMA) has been widely recognized as a promising way to scale up the number of users, enhance the spectral efficiency, and improve the user fairness in wireless networks, by allowing more than one user to share one wireless resource. NOMA can be flexibly combined with many existing wireless technologies and emerging ones including multiple-input multiple-output (MIMO), massive MIMO, millimeter wave communications, cognitive and cooperative communications, visible light communications, physical layer security, energy harvesting, wireless caching, and so on. Combination of NOMA with these technologies can further increase scalability, spectral efficiency, energy efficiency, and greenness of future communication networks. This paper provides a comprehensive survey of the interplay between NOMA and the above technologies. The emphasis is on how the above techniques can benefit from NOMA and vice versa. Moreover, challenges and future research directions are identified.

I. INTRODUCTION

NOMA addresses the user-density limits of orthogonal access by allowing simultaneous transmission over shared resources. The survey reviews NOMA's combination with emerging technologies and examines their mutual benefits.

  • Orthogonal access methods simplify receiver design but restrict the number of users to the number of orthogonal resources.
  • The survey reviews how NOMA combines with emerging technologies, emphasizing their interplay and mutual benefits.
  • NOMA allows multiple users to access the same wireless resources simultaneously, supporting massive connectivity in 5G and beyond.
  • NOMA uses superposition coding and successive interference cancellation to support concurrent downlink transmission for users with different channel gains.
  • NOMA research remains constrained by the need for channel state information and by unresolved challenges before commercial deployment.
  • Massive MIMO can obtain its highest spectral efficiency in underloaded systems with conventional linear processing, motivating its study alongside NOMA.

2) Millimeter wave communications:

The paper situates NOMA alongside millimeter-wave, cooperative, cognitive, physical-layer security, and energy-harvesting communications. These technologies target bandwidth, throughput, spectrum sharing, secrecy, and sustainable operation.

  • 2) Millimeter wave communications:: Millimeter-wave communications provide gigabit-per-second data rates through large bandwidths but support limited simultaneous connections because their channels are spatially sparse.
  • 3) Cooperative communications:: Cooperative communications share resources among network nodes, and NOMA can mutually support cooperation through relay-aided and multi-cell cooperative schemes.
  • 4) Cognitive radio:: Cognitive radio improves spectral efficiency through spectrum sharing, including concurrent operation with incumbent users and opportunistic use of unused spectrum.
  • 5) PHY security:: Physical-layer security complements higher-layer cryptography by exploiting channel noise, fading, and interference to protect confidential information from eavesdroppers.
  • 6) Energy harvesting communications:: Energy-harvesting communications aim to provide energy-self-sufficient operation and support greener communications by harvesting natural or ambient RF energy.

7) Visible light communications (VLC):

The survey situates NOMA within a broader set of wireless technologies and examines massive MIMO-NOMA design, performance, and practical constraints. For sub-6 GHz massive MIMO, reported gains depend strongly on CSI quality, interference, user loading, channel conditions, and transmission design.

  • The survey focuses on how NOMA and emerging technologies interplay, benefit from each other, and shape future wireless-network research.
  • Residual intra- and inter-cluster pilot contamination detrimentally affects overloaded NOMA performance even as the BS antenna count becomes asymptotically large.Perfect-CSI assumptions can therefore overestimate practical massive MIMO-NOMA performance.
  • NOMA outperforms OMA in achievable sum rate only with highly accurate CSI and no inter-cluster interference.Distinct path losses within clusters can boost NOMA’s gains, while more than one NOMA user-cluster favors OMA with multiuser spatial multiplexing.
  • NOMA performs better than multi-user massive MIMO when the BS antenna count and total user count are approximately equal, M ≈K.When M ≫K, massive MIMO-OMA with ZF precoding outperforms power-domain NOMA; NOMA gains are more prominent for LoS deterministic channels than NLoS i.i.d. Rayleigh fading.
  • Channel-covariance-based user classification, clustering, and pilot allocation outperform conventional non-orthogonal and orthogonal pilot allocation in achievable user rates.
  • Proper transmit power control is essential for boosting uplink massive MIMO-NOMA rates and ensuring fairness under near-far effects.A max-min optimal policy based on channel statistics is proposed when users lack downlink pilot transmissions.

B. Applications of Massive MIMO-NOMA in Relay Networks

The survey reviews NOMA-enabled massive MIMO relay networks and distributed-transmission systems as ways to extend NOMA performance beyond conventional centralized deployments. Reported designs use rate analysis, resource allocation, power optimization, and distributed antennas or access points.

  • Relay networks: The achievable sum rate increases linearly with the number of admitted relayed users in massive MIMO-NOMA relay networks.The ratio between transmit-antenna count and relay count also plays a key role in system-wide performance.
  • Relay networks: Efficient three-dimensional resource allocation can further improve NOMA-aided massive MIMO relaying performance.
  • Relay networks: Transmit power optimization at both the BS and relay nodes is important for spectral-efficiency gains in massive MIMO-NOMA relaying.
  • Distributed transmissions: Cell-free massive MIMO-NOMA derives downlink rates under beamforming uncertainty, residual interference, erroneous channel estimation, and imperfect SIC.Multi-antenna access points extend the single-antenna distributed-transmission model to leverage distributed multi-antenna transmission.
  • Distributed transmissions: User-centric distributed transmissions in cell-free underlay spectrum-sharing massive MIMO-NOMA improve secondary-system performance without hindering primary-system performance gains.

A. Design Insights and Implications

Research on mmWave massive MIMO-NOMA develops rate analyses, low-feedback and beamforming designs, lens-array architectures, and multi-beam extensions while highlighting angle-domain methods as a future direction.

  • Rate analysis: mmWave massive MIMO-NOMA rate bounds cover both noise-dominated low-SNR and interference-dominated regimes, supporting large spectral-efficiency gains.The bounds use deterministic-equivalent analysis with the Stieltjes-Shannon transform in the low-SNR regime.
  • Low-feedback design: A low-feedback design decomposes the massive MIMO-NOMA channel into SISO-NOMA channels, reducing computational complexity while balancing performance and implementation cost.The design assumes perfect user ordering and one-bit feedback.
  • Practical impairments: Finite-resolution analog precoders reduce hardware cost but create beam-alignment mismatches and received-power leakage in mmWave massive MIMO.The cited discussion contrasts the effects of imperfect alignment in OMA and NOMA systems.
  • Practical impairments: Hybrid beamforming studies derive sum-rate lower bounds under aligned LoS channels and model misaligned LoS or NLoS channels using a beam-misalignment factor.The resulting analog and digital precoders are designed to maximize sum rate.
  • Beamspace and multi-beam architectures: Lens antenna arrays let mmWave MIMO-NOMA serve more users simultaneously than the number of RF chains, while multi-beam designs address narrow-beam user limitations.Lens arrays transform spatial channels into the beamspace domain for rate analysis and precoder design.
  • Future directions: Prior NOMA-enabled massive MIMO work mainly exploits the spatial domain, motivating angle-domain pilot allocation, channel estimation, beamforming, power allocation, and interference mitigation.Deep learning is also identified as a tool for optimizing tradeoffs among channel estimation, power allocation, and iterative/SIC decoding.

IV. COEXISTENCE OF NOMA AND COOPERATIVE COMMUNICATIONS

The survey organizes NOMA-cooperative communications into cooperative NOMA, relay-aided NOMA, and multi-cell cooperative transmission, with reliability and cell-edge performance as recurring concerns.

  • Cooperative NOMA: Cooperative NOMA uses a strong-channel user as a decode-and-forward relay to forward decoded messages to a weak-channel user.The process uses two time slots: broadcast of superposed messages followed by relay forwarding.
  • Cooperative NOMA: Cooperative NOMA can improve weak-user reliability, fairness, redundancy, and diversity gain, but requires an extra transmission slot.Full-duplex relaying and relay-selection methods are discussed as ways to address or improve this tradeoff.
  • Relay-aided NOMA: Relay-aided NOMA research covers amplify-and-forward multi-antenna downlinks, Alamouti-based multi-cell uplinks, and coordinated direct-relay transmission.These works include outage-performance analysis for the downlink case.
  • Multi-cell cooperative transmission: Multi-cell NOMA addresses cell-edge users and intra- or inter-cell interference through network NOMA, CoMP, HetNets, and C-RAN architectures.CoMP enables multiple base stations to coordinate beamforming for cell-edge-user enhancement.
  • Discussions and outlook: Open issues include reconsidering SIC decoding order, coordinating power allocation, controlling error propagation, and reducing decoding hardware complexity.The near user need not have the best channel quality in multi-cell cooperative NOMA, and inappropriate power allocation can increase energy consumption.

A. NOMA in Cognitive Radio Networks

NOMA in cognitive radio networks enables secondary transmitters to serve multiple secondary users under primary-user interference constraints, while CR-inspired NOMA uses power allocation to protect weak-user QoS.

  • NOMA in cognitive radio networks: NOMA-CR allows a secondary transmitter to communicate with multiple secondary NOMA users while satisfying the primary-user interference-power constraint.The survey describes a two-user example and large-scale analyses using stochastic geometry.
  • Cognitive radio-inspired NOMA: CR-inspired NOMA treats the base station as combined primary and secondary transmitters and limits strong-user power to ensure weak-user QoS.The scheme applies the key feature of underlay cognitive radio through a novel power-allocation design.
  • Cognitive radio-inspired NOMA: CR-inspired NOMA can improve throughput-fairness flexibility and has been extended to MIMO-NOMA power allocation using signal alignment.
  • Discussions and outlook: Existing NOMA-CR studies mainly focus on underlay CR, leaving interweave and overlay NOMA-CR comparatively immature.Further performance gains may involve power allocation, user clustering, and user pairing.
  • Discussions and outlook: Future NOMA-CR research should address low-power massive connectivity for IoT and distributed resource management for dynamic UAV and V2X deployments.Centralized allocation may not suit dynamic UAV deployment because UAVs are expected to make local decisions.

VI. NOMA AND PHYSICAL LAYER SECURITY

NOMA physical-layer security combines established techniques such as artificial noise, beamforming, antenna selection, cooperative jamming, and relaying to protect against external and internal eavesdroppers. The surveyed literature derives secrecy metrics across MIMO, cooperative, stochastic-geometry, and VLC settings.

  • PHY security techniques exploit wireless-channel characteristics to protect NOMA transmissions from external and internal eavesdroppers.Approaches include artificial noise, beamforming, transmit antenna selection, cooperative jamming, and relay-based security.
  • Artificial-noise methods support secure NOMA when the eavesdropper’s CSI is unavailable, with secrecy-outage analyses developed for MISO and stochastic-geometry systems.For multi-antenna base stations, artificial noise is generated at the BS while legitimate users and eavesdroppers may be spatially random.
  • Cooperative relaying can produce strictly positive secrecy rates, whereas secrecy rates may be zero without relays; cooperative jamming is strongest when relays are near the eavesdropper.The preferred relaying scheme depends on node distances and the operating point in the secrecy-rate region.
  • Cooperative jamming lets relays forward confidential information while emitting jamming signals, and can also support secondary spectrum access under primary-user secrecy constraints.The surveyed schemes include two-way relay networks and overlay NOMA cognitive-radio systems.
  • Transmit antenna selection reduces RF-chain cost, complexity, size, and power consumption while retaining acceptable diversity and throughput benefits for secure MIMO-NOMA.The surveyed work derives secrecy-outage expressions and secrecy diversity results for TAS strategies.

D. Beamforming-based Strategies

The survey describes beamforming and energy-harvesting-assisted cooperative strategies for NOMA, emphasizing interference control, relay support, user pairing, and practical power consumption. These methods target achievable-rate, outage, and fairness improvements under specified network conditions.

  • D. Beamforming-based Strategies: ZF beamforming can place information in the eavesdropper’s channel null space when the transmitter has more antennas, while eliminating inter-user interference under perfect CSI.The approach is asymptotically optimal in massive MIMO settings according to the surveyed literature.
  • D. Beamforming-based Strategies: Beamforming-based PHY-security solutions remain concentrated on two-user NOMA systems, and extending them to large clusters with multiple users is identified as important future work.The survey also notes limited solutions for imperfect CSI and nontrivial extensions of beamforming approaches.
  • A. NOMA with SWIPT: SWIPT enables stronger downlink NOMA users to harvest energy from BS signals and forward data to weaker users.The cooperative model uses near users to assist far users through harvested energy.
  • B. NOMA with WPCN: WPCN systems first transfer energy wirelessly from the BS, after which users transmit uplink data using NOMA.The surveyed optimization selects energy-transfer and transmission durations, decoding orders, and fairness or sum-rate objectives.
  • A. NOMA with SWIPT: NNNF pairing minimizes outage probability and maximizes achievable rates for both near and far users in the surveyed SWIPT model.Closed-form performance results are obtained, and suitable transmission rates and power-splitting coefficients can provide guaranteed performance without using near users’ own energy for relay transmission.

B. NOMA with WPCN

NOMA with WPCN studies sequential wireless energy transfer and uplink information transmission under optimization objectives for sum rate and fairness. Results depend critically on whether circuitry power consumption is included.

  • WPCN-NOMA optimization chooses the energy-transfer duration and decoding orders to maximize total sum rate or minimum user rate.The considered optimization problems are linear or convex, facilitating practical optimal solutions.
  • NOMA with WPCN increases user fairness compared with conventional OMA when uplink transmissions are constrained by harvested energy.This conclusion comes from comparing the two access methods under the harvested-energy constraint.
  • When circuitry power consumption pc > 0, TDMA outperforms NOMA in both energy efficiency and spectral efficiency in the studied uplink WPCN setting.The survey therefore stresses accurate power-consumption modeling before selecting a transmission scheme for energy-constrained devices.
  • Extending NOMA energy-harvesting studies to MIMO requires accounting for the tradeoff between multiple-antenna gains and additional circuitry power.The survey also identifies nonlinear received-energy models and interference harvesting as directions for future NOMA research.

A. Existing Literature of NOMA-VLC

NOMA-VLC research addresses user grouping, power allocation, fairness, peak-power constraints, mobility, and practical SIC effects. The distinctive LED amplitude constraints make signaling design a central open problem.

  • With residual SIC interference, NOMA-VLC has a superior achievable rate region to conventional OFDMA in the surveyed two-user downlink comparison.VLC positioning is also used to group users, with SIC assigned according to relative channel conditions within groups.
  • Gain-ratio power allocation assigns power according to channel quality to improve fairness for low-decoding-order users that experience large interference.The effect of LED transmission angles is also studied.
  • VLC-NOMA studies derive outage expressions for guaranteed QoS and ergodic sum-rate expressions for opportunistic best-effort service, while accounting for LED lighting characteristics.
  • MIMO-VLC-NOMA simulations show higher achievable sum rates for normalized-gain-difference power allocation than for a generalized gain-ratio strategy.Detection uses zero forcing followed by SIC.
  • Under LED peak-power constraints, proposed power-control algorithms outperform other allocation strategies after non-convex fairness problems are converted to convex form.These constraints avoid clipping distortion, and imperfect CSI effects are also analyzed through closed-form bit-error-rate results.
  • Future VLC-NOMA signaling must address LED amplitude constraints because Gaussian signaling is infeasible and therefore cannot be assumed optimal.The constraints preserve the LEDs’ dynamic range and avoid clipping distortion.

IX. NOMA WITH MOBILE EDGE COMPUTING

NOMA can be integrated with MEC in both uplink and downlink so multiple users offload tasks simultaneously, targeting massive connectivity, low latency, and high spectral efficiency. Research focuses on jointly optimizing communication and computing resources, while imperfect CSI remains a key practical obstacle.

  • NOMA-MEC supports simultaneous task offloading by multiple users in the same frequency band for both uplink and downlink transmission.This combination is associated with massive connectivity, low latency, and high spectral efficiency.
  • NOMA-MEC can provide lower latency and lower energy consumption than traditional OMA, but resource optimization is central to achieving these benefits.Relevant variables include offloading power, bandwidth, offloading time, task assignment, and computing resource blocks.
  • Partial, binary, and fully offloading schemes are studied alongside energy-efficient designs and task-delay minimization.The surveyed work includes weighted sum-energy minimization and hybrid NOMA-MEC systems for fully offloaded tasks.
  • Joint allocation of transmit power, subcarriers, computing capacity, and CPU frequency is identified as important for reducing task delay and energy consumption.The relevant optimization differs between uplink and downlink MEC configurations but spans both communication and computation resources.
  • Existing NOMA-MEC studies mainly assume perfect CSI, whereas channel-estimation errors, partial CSI, and limited feedback can complicate SIC decoding order.Improving performance under imperfect CSI is identified as an important research direction.

3) Security in NOMA-MEC:

NOMA-MEC raises secrecy and privacy concerns because eavesdroppers may attempt to decode users’ messages. The surveyed research places this issue alongside cooperative extensions, broader NOMA integrations, and learning-based approaches for difficult optimization problems.

  • 3) Security in NOMA-MEC:: Passive or active eavesdroppers may attempt to decode mobile users’ messages during NOMA-MEC offloading, motivating physical-layer security solutions.The cited security approach addresses scenarios with external eavesdroppers.
  • 3) Security in NOMA-MEC:: Cooperative NOMA-MEC can improve connectivity for distant devices by using a nearby mobile device as a relay to help offload tasks.The cooperating device transmits superimposed signals to the primary MEC server and the helper relay.
  • 3) Security in NOMA-MEC:: NOMA has also been integrated with vehicular, terrestrial-satellite, UAV, ambient-backscatter, wireless-caching, and Wi-Fi communications.Common challenges across these integrations include clustering, power allocation, and SIC performance.
  • 3) Security in NOMA-MEC:: Machine learning and deep learning are used to address large-scale and NP-hard uplink and downlink NOMA optimization problems.One described deep-learning workflow uses exploration, training, and exploitation for power allocation in caching-based NOMA.
  • 3) Security in NOMA-MEC:: The survey concludes that combining NOMA with emerging technologies can overcome limits that those technologies cannot overcome singlehandedly, while leaving challenges and future directions open.The conclusion covers massive MIMO, mmWave, cooperative communications, physical-layer security, visible light, energy harvesting, MEC, and machine learning.
Loading 1903.10489v2…