PaperlayerView exact revision on arXiv

Source-linked AI summary

Rate-Splitting Multiple Access for 6G -- Part I: Principles, Applications and Future Works

Anup Mishra, Yijie Mao, Onur Dizdar, Bruno Clerckx

arXiv:2205.02548v3cs.ITeess.SPmath.IT

TL;DR

6G's diverse service requirements pose challenges for existing multiple-access schemes. This tutorial explains RSMA's principles and applications for 6G, and concludes that it is a promising NGMA scheme for future networks.

  • Problem

    6G's massive-access needs and diverse services challenge conventional multiple-access schemes, whose limitations include sensitivity to CSIT impairments and restrictions on interference management.

    Source: §I · pp. 1–2

  • Method

    The letter tutorials RSMA's design principles and downlink and uplink frameworks, then surveys applications and future work across 6G technologies and use cases.

    Source: §V · p. 5

  • Results

    The tutorial concludes that RSMA is a promising next-generation multiple-access scheme for future networks such as 6G and beyond.

    Source: §V · p. 5

  • Takeaways & Limitations

    RSMA can help address interference and access challenges in optical wireless communications, waveform design, network slicing, and integration with other multiple-access schemes.

    Source: 3) Optical wireless communications (OWC) · p. 4, 4) Waveform design · p. 4, 6) Network slicing · p. 4

    Users must know message splitting and combining, which increases signaling burden, while splitting and combining also complicate HARQ.

    Source: §IV · p. 5

Abstract

from arXiv · show

This letter is the first part of a three-part tutorial focusing on rate-splitting multiple access (RSMA) for 6G. As Part I of the tutorial, the letter presents the basics of RSMA and its applications in light of 6G. To begin with, we first delineate the design principle and basic transmission frameworks of downlink and uplink RSMA. We then illustrate the applications of RSMA for addressing the challenges of various potential enabling technologies and use cases, consequently making it a promising next generation multiple access (NGMA) scheme for future networks such as 6G and beyond. We briefly discuss the challenges of RSMA and conclude the letter. In continuation of Part I, we will focus on the interplay of RSMA with integrated sensing and communication, and reconfigurable intelligent surfaces, respectively in Part II and Part III of this tutorial.

I. INTRODUCTION

6G's demanding connectivity and service goals make multiple-access design a central challenge. The introduction positions RSMA as an approach to limitations of existing schemes and outlines the letter's tutorial scope.

I. INTRODUCTION

  • 6G's diverse services and emerging functions require massive access and high quality of service, making multiple-access design a formidable challenge.The passage cites targets of 50 times higher peak data rate, 10 times reduced latency, and 10^4 times higher reliability than 5G.Source: §I · p. 1
  • OMA avoids inter-user interference but uses radio resources inefficiently, while SDMA shares time-frequency resources but can retain multi-user interference.SDMA uses precoders or spatial beams to subdue interference relative to noise.Source: §I · p. 1
  • PD-NOMA uses SIC to decode and remove interference, but full interference decoding limits its downlink advantages in multi-antenna settings and increases receiver complexity.In the uplink, PD-NOMA relies on time sharing, adding communication overhead and strict synchronization requirements.Source: §I · p. 1
  • RSMA splits messages, superposes their parts, and manages interference across power and spatial domains by adjusting the split and power allocation.The introduction reports optimal downlink DoF with perfect and imperfect CSIT and the uplink capacity region without user time sharing.Source: §I · p. 2
  • The letter introduces RSMA frameworks, surveys applications and future work for 6G technologies and use cases, and briefly discusses its challenges.The stated sequence covers design principles and transmission frameworks, applications, then challenges and conclusions.Source: §I · p. 2

II. RSMA: DESIGN PRINCIPLE AND TRANSMISSION FRAMEWORK

RSMA manages interference by partially decoding it and partially treating it as noise; its uplink and downlink motivations differ. The downlink framework splits each user's message into common and private parts, then decodes and recombines their streams.

II. RSMA: DESIGN PRINCIPLE AND TRANSMISSION FRAMEWORK

  • RSMA partially decodes interference and partially treats it as noise, with uplink use motivated by avoiding time sharing and downlink use by improving SE.This design generalizes PD-NOMA and SDMA's interference-management strategies.Source: §II · p. 2

A. RSMA in the downlink

  • In the one-layer downlink framework, users' common message parts are combined into a common stream, while each private part is separately encoded.Each user first decodes the common stream while treating private streams as noise, then uses SIC to remove it before decoding its own private stream.Source: A. RSMA in the downlink · p. 2
  • Each user's total rate combines its private-stream rate with its share of the common-stream rate, and precoders may be designed with low-complexity methods or optimized for objectives such as weighted sum-rate or EE.The user reconstructs its message by combining its common and private parts.Source: A. RSMA in the downlink · p. 3
  • RSMA is reported to outperform OMA, NOMA, and SDMA in DoF, SE, and EE under perfect and imperfect CSIT and underloaded and overloaded network loads.The passage also reports practical throughput gains in link-level simulations and multi-layer RSMA gains over SDMA and NOMA in SE with imperfect CSIT.Source: A. RSMA in the downlink · p. 3

B. RSMA in the uplink

Uplink RSMA splits the messages of K−1 users into two streams while leaving user K unsplit, then uses receiver-side decoding. It can achieve the Gaussian MAC capacity region without time sharing, at the cost of multiple SIC layers.

B. RSMA in the uplink

  • K−1 users split their messages into two independently encoded streams, while user K transmits one directly encoded stream.The streams are power-allocated and superposed; Figure 2 depicts the K-user uplink transceiver architecture.Source: B. RSMA in the uplink · p. 3, Figure 2 · p. 3
  • The MISO downlink transceiver architecture can be extended to MIMO by splitting and encoding vectors of user messages into vectors of common and private streams.This extension is noted for the architecture referenced in the uplink discussion.Source: B. RSMA in the uplink · p. 3
  • The base station uses receive filters and a decoding order for the 2K−1 streams, with user rates depending on both decoding order and transmit power allocation.The decoding order and powers are optimized for uplink sum-rate performance.Source: B. RSMA in the uplink · p. 3
  • Uplink RSMA achieves the Gaussian MAC capacity region without time sharing, but requires multiple SIC layers at the base station.It can support users with intermittent access, including URLLC and mMTC.Source: B. RSMA in the uplink · p. 3

III. APPLICATIONS AND FUTURE WORKS

RSMA's potential applications and future work are considered in relation to enabling technologies and core services envisioned for 6G.

III. APPLICATIONS AND FUTURE WORKS

  • RSMA's emerging applications and potential future work are considered for 6G enabling technologies and core services.The letter frames these as applications of RSMA in multi-antenna communications.Source: §III · p. 4

1) Massive antennas:

In massive-antenna networks, inaccurate CSIT limits performance, while RSMA is more robust to CSIT inaccuracies than SDMA.

1) Massive antennas:

  • RSMA is more robust than SDMA to CSIT inaccuracies in massive-antenna networks and can benefit RIS and lens antenna arrays, which are also vulnerable to such inaccuracies.Named sources of CSIT inaccuracy include mobility, hardware impairments, and pilot contamination.Source: 1) Massive antennas · p. 4

2) Ultra high frequencies:

In mmWave communications, RSMA is reported to outperform SDMA in sum rate while requiring less CSIT acquisition complexity; the letter also identifies potential applications in THz communications.

2) Ultra high frequencies:

  • RSMA achieves higher sum-rate performance and lower CSIT acquisition complexity than SDMA in mmWave communications.The letter identifies inefficient user synchronization and low-resolution analog-to-digital converters as THz challenges RSMA may help address.Source: 2) Ultra high frequencies · p. 4

3) Optical wireless communications (OWC):

RSMA is considered for optical wireless communication, including VLC, where it is reported to provide gains despite channel correlation and imperfect CSI.

3) Optical wireless communications (OWC):

4) Waveform design:

The section surveys RSMA applications in waveform design, integration with other multiple-access schemes, and network slicing. It describes potential interference-management benefits and reports an URLLC advantage in the presence of eMBB interference.

4) Waveform design:

  • RSMA can support flexible waveform design by addressing inter-channel interference, asynchronous transmissions, processing delays, Doppler effects, and reception under imperfections.The passage also identifies potential integration with vortex waves and waveform design for ISAC.Source: 4) Waveform design · p. 4

5) Integration with other MA schemes:

6) Network slicing:

  • RSMA outperforms OMA and NOMA for URLLC service in network slicing, even with interference from eMBB users.Source: 6) Network slicing · p. 4

7) Distributed networks:

The section describes RSMA applications in distributed networks, mMTC, V2X, and ISAC. It reports performance benefits in several settings while identifying open challenges and research directions.

7) Distributed networks:

  • RSMA’s interference-management capabilities may help distributed networks address imperfect CSIT, congestion, and resource allocation, with potential throughput, EE, and security benefits.Source: 7) Distributed networks · p. 4

8) mMTC:

  • In mMTC, RSMA improves uplink reliability and achieves better outage performance than NOMA; downlink SE gains over SDMA are reported with deteriorated CSIT in cell-free MaMIMO.The coexistence of mMTC with URLLC and eMBB remains to be investigated.Source: 8) mMTC · p. 4

9) Vehicular to everything (V2X) communications:

  • For V2X, RSMA is described as more robust to mobility and as providing rate benefits with shorter blocklengths; dense-environment collisions and spectrum-sharing interference remain challenges.Source: 9) Vehicular to everything (V2X)… · p. 4

10) ISAC:

  • RSMA-aided DFRC improves both sensing and communication performance and reduces hardware complexity compared with SDMA-aided DFRC.Waveform design and security are identified as possible next steps for RSMA in ISAC.Source: 10) ISAC · p. 4

11) Artificial intelligence and machine learning (ML):

The section reviews RSMA applications with machine learning, intelligent edge computing, and space-air-ground integrated networks. It reports comparisons in specific settings and outlines potential integrations and challenges.

11) Artificial intelligence and machine learning (ML):

  • RSMA with ML has been used for power control without prior CSI information, achieving rate-performance gains over an SDMA-assisted network; ML integration also improved UAV deployment and reduced transmit power versus OMA.The passage recommends extending RSMA–ML integration beyond the PHY layer to cross-layer design, including network orchestration.Source: 11) Artificial intelligence and… · p. 4

12) Intelligent edge computing:

  • In cognitive-radio-network task offloading, RSMA has been shown to aid MEC better than NOMA.Integrating RSMA and ML with MEC is proposed for intelligent edge computing, including network orchestration and potentially improved fairness, performance, and security.Source: 12) Intelligent edge computing · pp. 4–5

13) Space-air-ground integrated networks (SAGIN):

  • In satellite and UAV integrated networks, RSMA achieves superior SE and EE performance to current MA schemes; it may help SAGIN address mobility, spectrum sharing, interoperability, and heterogeneous QoS constraints.Source: 13) Space-air-ground integrated… · p. 5

14) Simultaneous wireless information and power transfer (SWIPT):

RSMA-assisted SWIPT is reported to improve rate performance over SDMA under a specified energy constraint, while the tutorial identifies unresolved design and implementation challenges and presents RSMA as a potential NGMA for 6G and beyond.

14) Simultaneous wireless information and power transfer (SWIPT):

  • RSMA-assisted SWIPT achieves rate gains over SDMA under a fixed sum-energy constraint and saves more transmit power for co-located energy and information receivers.Future RSMA-SWIPT integration with RIS is proposed as a way to address limited transmission range to energy receivers and benefit low-powered-device services.Source: 14) Simultaneous wireless information… · p. 5

IV. CHALLENGES

  • Downlink SIC alternatives and low-complexity uplink message-splitting and power-allocation methods remain research challenges for reducing receiver complexity and signalling overhead.Source: §IV · p. 5
  • RSMA implementation still requires addressing signalling burden at both ends, HARQ complexity, un explored cross-layer and CSI-feedback design, and prototype validation.Users need message-splitting knowledge to decode and remove the common stream; HARQ must handle retransmissions after incorrect decoding.Source: §IV · p. 5

V. CONCLUSIONS

  • The letter presents RSMA’s principles, frameworks, applications, and future works, with later tutorial parts covering ISAC and RIS interplay.Source: §V · p. 5
Loading 2205.02548v3…