Source-linked AI summary
Multiple Antenna Assisted Non-Orthogonal Multiple Access
Yuanwei Liu, Hong Xing, Cunhua Pan, Arumugam Nallanathan, Maged Elkashlan, Lajos Hanzo
TL;DR
The article examines how multiple antennas shape NOMA rate regions and surveys multiple-antenna designs, resource-allocation problems, and implementation constraints. It reports rate-region gains over MIMO-OMA and theoretical optimality in important special cases.
Problem
The article addresses the rate-region behavior of multiple-antenna NOMA and the resource-allocation and implementation issues associated with these systems.
Source: §I · p. 2Method
It analyzes MIMO-NOMA rate regions, reviews two multiple-antenna NOMA designs, and discusses resource-allocation problems and implementation issues.
Source: §I · p. 2Results
MIMO-NOMA is reported to improve downlink spectral efficiency, with rate-region optimality in important special cases and gains over MIMO-OMA demonstrated theoretically.
Source: §II · p. 2, §VII · p. 10Takeaways & Limitations
The article identifies implementation solutions and open research directions, including SIC decoding order and spatial-effect investigation.
Source: §VII · p. 10
Abstract
from arXiv · showhide
Non-orthogonal multiple access (NOMA) is potentially capable of circumventing the limitations of the classic orthogonal multiple access schemes, hence it has recently received significant research attention both in industry and academia. This article is focused on exploiting multiple antenna techniques in NOMA networks, with an emphasis on investigating the rate region of multiple-input multiple-output (MIMO)-NOMA, whist reviewing two popular multiple antennas aided NOMA structures, as well as underlining resource management problems of both single-carrier and multi-carrier MIMO-NOMA networks. This article also points out several effective methods of tackling the practical implementation constraints of multiple antenna NOMA networks. Finally, some promising open research directions are provided in context of multiple antenna aided NOMA.
I. INTRODUCTION
The article examines how multiple antennas affect NOMA rate regions and surveys associated designs, resource-allocation problems, and implementation issues. It reports that MIMO-NOMA can improve downlink spectral efficiency and is rate-region optimal in important special cases.
I. INTRODUCTION
- The article frames multiple-antenna NOMA as a response to rising bandwidth demand from high-rate services and growing IoT connectivity needs.NOMA is presented as potentially improving spectral efficiency while supporting many devices.Source: §I · p. 1
- Multiple antennas can provide array gains through beamforming, increase throughput through spatial multiplexing, and create user-specific channels that remove NOMA channel-difference constraints.The resulting generalized NOMA design can address heterogeneous user QoS requirements.Source: §I · p. 2
- The article reviews two multiple-antenna NOMA designs and discusses their resource-allocation problems for both single-carrier and multi-carrier systems.It also identifies practical implementation issues and promising solutions.Source: §I · p. 2
II. MULTIPLE-ANTENNA NOMA: INFORMATION THEORETIC PERSPECTIVE
- MIMO-NOMA can improve downlink spectral efficiency and is theoretically rate-region optimal in important special cases.The article presents rate-region gains over MIMO-OMA from an information-theoretic perspective.Source: §II · p. 2, §I · p. 2
A. MISO-NOMA
For MISO-NOMA, the rate region under a fixed SIC order is generally contained within the DPC rate region, though NOMA can match DPC in special cases. For MIMO-NOMA, equality of NOMA, DPC, and capacity regions is established for aligned degraded channels, while the general capacity region remains unknown.
B. MIMO-NOMA
- The capacity region is unknown for general MISO and MIMO downlink broadcast channels, although DPC achieves capacity in several special MIMO cases.The MISO passage notes this uncertainty even for two users with one receive antenna each.Source: A. MISO-NOMA · p. 3, B. MIMO-NOMA · p. 3
A. MISO-NOMA
- For a fixed two-user MISO decoding order, the NOMA rate region is contained in the DPC rate region because SIC imposes an additional decodability constraint.NOMA requires the stronger receiver to decode the other user's message before cancelling it.Source: A. MISO-NOMA · p. 3
- Natural ordering by channel gains does not necessarily satisfy the rate condition required for successful SIC in MISO-NOMA.The required condition is Rn→m > Rm→m.Source: A. MISO-NOMA · p. 3
B. MIMO-NOMA
- For an aligned degraded MIMO broadcast channel, the capacity region, DPC rate region, and NOMA rate region coincide.The passage describes this equality under a covariance-matrix input constraint.Source: B. MIMO-NOMA · p. 3
III. MULTIPLE-ANTENNA NOMA: BEAMFORMER BASED STRUCTURE
The paper distinguishes beamformer-based NOMA, where a beamformer serves one user, from cluster-based NOMA. It then describes centralized beamforming and its decoding constraints.
A. Centralized Beamforming
- Centralized beamforming uses a single base station to superpose two users’ messages with a separate beamformer for each user.The example has the base station transmit messages for Users m and n.Source: A. Centralized Beamforming · p. 4
- User n first decodes User m’s message, then applies SIC and decodes its own message if decoding succeeds.The decoding rates are Rn→m and Rn→n, respectively.Source: A. Centralized Beamforming · p. 4
- Successful SIC requires Rn→m > Rm→m, making optimization challenging, while finding the optimal MISO-NOMA ordering remains open.Existing approaches may assume different path losses or a predefined decoding order.Source: A. Centralized Beamforming · p. 4
B. Coordinated Beamforming
Coordinated beamforming combines transmissions from multiple base stations to serve a cell-edge user. Its described benefits are paired with the capacity cost of inter-node information exchange.
B. Coordinated Beamforming
- Cooperation among single-antenna devices forms virtual antenna arrays, but requires setting aside achievable capacity for inter-node information exchange.The paper relates the approach to coordinated multi-point transmission.Source: B. Coordinated Beamforming · p. 5
- Several base stations can serve a cell-edge user while each also serves one user in its own cell, using SIC to cancel interference from the cell-edge user.This arrangement is described as enhancing cell-edge performance and improving network fairness.Source: B. Coordinated Beamforming · p. 5
- Two coordinated base stations used Alamouti-code-based coordinated SC to serve nearby user pairs and a cell-edge user; a later multiple-antenna extension considered BSs and users in a coordinated MIMO-NOMA setup.The extension developed a joint centralized and coordinated beamforming design.Source: B. Coordinated Beamforming · p. 5
IV. MULTIPLE ANTENNA NOMA: CLUSTER BASED STRUCTURE
The cluster-based structure groups users sharing beamformers, with designs that either eliminate or tolerate inter-cluster interference. The section outlines their mechanisms, constraints, and fairness-related resource allocation.
A. Inter-Cluster Interference Free Design
- The interference-free design decomposes MIMO-NOMA into SISO-NOMA channels and uses zero-forcing detection, avoiding sophisticated channel ordering and canceling inter-cluster interference.The decomposition is presented as reducing system complexity.Source: A. Inter-Cluster Interference Free… · pp. 5–6
- The adopted detection method requires a specific relationship between transmitter and receiver antenna counts: either N ≥ M [7] or N ≥ M/2 must hold.The passage notes that fairness is important in clustered NOMA and describes resource allocation for max-min fairness in clustered MIMO-NOMA with heuristic user scheduling.Source: A. Inter-Cluster Interference Free… · p. 6
B. Inter-Cluster Interference Tolerant Design
- The interference-tolerant design uses decoding scaling weights to strengthen desired signals despite inter-cluster interference.Its user clustering sorts users by equivalent normalized channel gain, with the highest-gain user as cluster-head, which can cancel intra-cluster interference using SIC.Source: B. Inter-Cluster Interference Tolerant… · p. 6
- The interference-tolerant design imposes no constraints on the base station or users’ antenna counts.The passage says this permits extension to massive-MIMO-NOMA and millimeter-wave scenarios.Source: B. Inter-Cluster Interference Tolerant… · p. 6
V. RESOURCE ALLOCATION FOR MULTIPLE-ANTENNA NOMA
Resource allocation in multi-antenna NOMA seeks spectral and energy efficiency through resource reuse, beamforming, user scheduling, and power allocation. The single-carrier discussion highlights joint clustering, assignment, and power decisions.
V. RESOURCE ALLOCATION FOR MULTIPLE-ANTENNA NOMA
- Multi-antenna NOMA resource allocation seeks to enhance spectral and energy efficiency through resource reuse, beamforming, user scheduling, and power allocation.The section covers both single-carrier and multi-carrier scenarios and introduces mathematical modeling tools.Source: §V · p. 7
A. Single-Carrier Resource Allocation
- Optimally solving the single-carrier allocation problems remains challenging and is identified as a research direction.The discussion includes throughput and max-min fairness objectives under particular transmission-power control and decoder choices.Source: A. Single-Carrier Resource Allocation · p. 7
- Single-carrier clustered MIMO-NOMA must determine the number and size of clusters, user-to-cluster assignments, and power sharing within and across clusters.The setup uses a single carrier and the same time slot or spreading code.Source: A. Single-Carrier Resource Allocation · p. 7
- Some studies simplify single-carrier allocation by fixing cluster number and size, then optimizing user scheduling and power allocation.Those studies focus on only two of the listed allocation issues.Source: A. Single-Carrier Resource Allocation · p. 7
B. Multi-Carrier Resource Allocation
Multi-carrier MIMO-NOMA adds frequency-domain multiplexing to power and spatial multiplexing, making resource allocation a multidimensional problem. The section outlines its design complexity and optimization approaches for single- and multi-carrier MIMO-NOMA.
B. Multi-Carrier Resource Allocation
- Multi-carrier MIMO-NOMA combines frequency-domain multiplexing with power and spatial multiplexing to manage multidimensional resources.The design entails two rounds of user scheduling and power allocation across sub-carriers and MIMO clusters.Source: B. Multi-Carrier Resource Allocation · pp. 7–8
- Partitioning users into orthogonal sub-carrier bands and applying MIMO-NOMA within each band can significantly reduce implementation complexity.The design must manage interference as the number of users per resource block grows, which may require more antennas at the base station.Source: B. Multi-Carrier Resource Allocation · pp. 7–8
- Multi-carrier MIMO-NOMA designs remain in their infancy, while initial research results are available for multi-carrier SISO-NOMA.The associated resource allocation requires more sophisticated design than the single-carrier extension alone suggests.Source: B. Multi-Carrier Resource Allocation · pp. 7–8
C. Approaches for Resource Allocation
- MIMO-NOMA resource-allocation formulations are usually mixed-integer non-convex problems, addressed through joint or decoupled optimization.Joint user scheduling and power allocation can use monotonic optimization or Branch-and-Bound, which may yield globally optimal solutions.Source: C. Approaches for Resource Allocation · p. 8
- Decoupling scheduling from power allocation offers moderate-complexity matching for scheduling and geometric programming or non-cooperative games for power allocation.The cited decoupled approaches are described as optimization of two separate sub-problems.Source: C. Approaches for Resource Allocation · p. 8
VI. TACKLING THE PRACTICAL IMPLEMENTATION CONSTRAINTS OF MULTIPLE-ANTENNA NOMA
Multiple-antenna NOMA may improve network performance as antenna counts scale, but practical systems face overhead, design, energy, and security constraints. The section presents potential approaches including imperfect-CSI designs, wireless power transfer, and antenna selection.
VI. TACKLING THE PRACTICAL IMPLEMENTATION CONSTRAINTS OF MULTIPLE-ANTENNA NOMA
- Multiple-antenna NOMA can impose feedback overhead and sophisticated TPC and detector requirements, alongside energy and security issues.The section identifies effective solutions for these implementation constraints.Source: §VI · p. 8
A. Reducing Complexity with Imperfect CSI
- Because perfect CSI is unattainable in practical systems, low-complexity imperfect-CSI designs are desired for MIMO-NOMA.The approaches described use partial CSI, such as slowly fluctuating path loss, or limited feedback such as a TPC codebook index to reduce overhead.Source: A. Reducing Complexity with Imperfect… · p. 9
B. Tackling Energy Issues with Multi-Antenna Aided Wireless Power Transfer
- Multi-antenna techniques support SWIPT in NOMA networks, offering opportunities for energy-constrained IoT devices but requiring further research on joint energy and information beam design.The passage notes that user equipment in wireless sensor networks is usually energy constrained.Source: B. Tackling Energy Issues with… · p. 9
C. Low-Complexity Design with Antenna Selection
- Antenna selection can reduce radio-frequency-chain hardware costs without losing spatial diversity and can transform MIMO-NOMA into SISO-NOMA, avoiding sophisticated channel ordering.Designing effective transmitter and receiver antenna-selection algorithms to balance performance and complexity remains a promising direction.Source: C. Low-Complexity Design with Antenna… · p. 9
D. Security Provision Relying on Multiple Antennas
The section considers physical-layer security risks from wireless broadcasting and successive interference cancellation in NOMA. It describes multi-antenna approaches to security while noting an unresolved internal-user eavesdropping problem.
D. Security Provision Relying on Multiple Antennas
- Wireless broadcasting and SIC create security threats in NOMA, motivating physical-layer security measures.Source: D. Security Provision Relying on… · p. 9
- Multiple antennas can enhance physical-layer security by increasing the desired user's capacity or directing artificial noise toward eavesdroppers without degrading the desired user's reception.The artificial-noise approach is described for a MISO-NOMA scenario with external eavesdroppers.Source: D. Security Provision Relying on… · pp. 9–10
- Preventing a poor-channel NOMA user from detecting a better-channel user's message remains an open problem.The section discusses using multiple antennas to create effective channel differences between the users.Source: D. Security Provision Relying on… · p. 10
VII. CONCLUSIONS AND PROMISING RESEARCH DIRECTIONS
The article reviews multiple-antenna NOMA capacity, structures, resource allocation, and implementation constraints, then identifies open research directions. These include spatial effects, decoding order, modulation, and integration with massive MIMO and millimeter-wave systems.
VII. CONCLUSIONS AND PROMISING RESEARCH DIRECTIONS
- The information-theoretic gain of MIMO-NOMA over MIMO-OMA has been demonstrated.Source: §VII · p. 10
- The article highlights beamformer-based and cluster-based NOMA structures, resource-allocation problems, implementation issues, and potential solutions.Source: §VII · p. 10
- Spatial-effect research can extend stochastic-geometry analysis from BPP and PPP models toward the more practical but challenging PCP, incorporating directional beamforming and user locations.The article identifies spatial effects as an open research problem.Source: §VII · p. 10
- Optimal MIMO-NOMA SIC decoding order remains open because ordering depends on TPC and detector design, while existing work often assumes a particular ordering that does not approach the optimal performance bound.Source: §VII · p. 10
- MIMO-NOMA modulation design is an open direction as these systems are expected to be incorporated into OFDM and other novel modulation schemes.Source: §VII · pp. 10–11
- Coexistence of NOMA with massive MIMO and millimeter wave could improve spectral efficiency and support massive connectivity, but requires redesign of TPC and detection techniques.Source: §VII · p. 11
IEEE Communications Letters and the IEEE Access.
The supplied section combines author biographies with labels for rate-region, multiple-antenna NOMA structure, and resource-allocation visuals.
IEEE Communications Letters and the IEEE Access.
- Cunhua Pan's biography lists research interests in ultra-dense C-RAN, UAV, IoT, NOMA, and mobile edge computing.It notes his research-fellow role at Queen Mary University of London.Source: IEEE Communications Letters and the… · p. 12
- Arumugam Nallanathan's biography notes his wireless-communications role and interests including 5G, the Internet of Things, and molecular communications.It also records his Best Paper Award at IEEE ICC 2016.Source: IEEE Communications Letters and the… · p. 12
- The visuals are titled as covering NOMA and other schemes' rate-region relationships and representative MIMO-NOMA resource-allocation optimization approaches.The supplied captions identify their scope but do not provide the comparisons or optimization findings.Source: Table I · p. 13, Table II · p. 14
- The remaining diagrams are titled as illustrating beamformer-based and cluster-based multiple-antenna NOMA structures and physical-layer security.The supplied captions do not describe the diagrams' internal encodings.Source: Figure 2 · p. 14, Figure 3 · p. 15, Figure 4 · p. 16