Uplink rate-splitting multiple access for 6G
File(s)
Author(s)
Xu, Jiawei
Type
Thesis
Abstract
Rate-Splitting Multiple Access (RSMA) is a pioneering framework that optimises non-orthogonal transmission, Multiple Access, and interference management in next-generation wireless networks. This thesis assesses uplink RSMA performance for both Single-Input Single-Output (SISO) and Multiple-Input Multiple-Output (MIMO) channels in the Finite Blocklength (FBL) regime, and designs the Physical (PHY)-layer architecture for uplink MIMO RSMA. Additionally, it explores the implementation of RSMA in Mobile Edge Computing (MEC) and evaluates various scenarios.
First, the performance of uplink SISO RSMA is investigated under FBL constraints with and without time-sharing. Sum-throughput and rate region maximisation problems by optimising power and rate allocations are formulated and solved using the Successive Convex Approximation (SCA)-based algorithm. The impact of blocklength and transmission rate on the throughput and error probability performance of RSMA are investigated. According to simulation results, RSMA is demonstrated to be promising for FBL communications.
Next, the research scope is extended from uplink SISO RSMA in the FBL regime to uplink MIMO RSMA. An uplink MIMO RSMA framework is proposed and a problem optimising both precoders and combiners with Max-Min Fairness (MMF) metric is investigated. The Alternating Optimisation (AO) is applied to decompose the problem into two subproblems due to the high coupling between precoders and combiners. The PHY-layer uplink MIMO RSMA architecture is designed. The effectiveness and robustness of the proposed uplink MIMO RSMA schemes are demonstrated through numerical results.
Last, the application of RSMA with the framework of MEC in short-packet communication is examined. A successful computation probability maximisation problem is formulated by optimising the offloading factor, power allocation and task-splitting factor. The optimisation problem is solved by SCA-based algorithm. The results indicate that RSMA is reliable for communication in MEC under FBL constraints.
First, the performance of uplink SISO RSMA is investigated under FBL constraints with and without time-sharing. Sum-throughput and rate region maximisation problems by optimising power and rate allocations are formulated and solved using the Successive Convex Approximation (SCA)-based algorithm. The impact of blocklength and transmission rate on the throughput and error probability performance of RSMA are investigated. According to simulation results, RSMA is demonstrated to be promising for FBL communications.
Next, the research scope is extended from uplink SISO RSMA in the FBL regime to uplink MIMO RSMA. An uplink MIMO RSMA framework is proposed and a problem optimising both precoders and combiners with Max-Min Fairness (MMF) metric is investigated. The Alternating Optimisation (AO) is applied to decompose the problem into two subproblems due to the high coupling between precoders and combiners. The PHY-layer uplink MIMO RSMA architecture is designed. The effectiveness and robustness of the proposed uplink MIMO RSMA schemes are demonstrated through numerical results.
Last, the application of RSMA with the framework of MEC in short-packet communication is examined. A successful computation probability maximisation problem is formulated by optimising the offloading factor, power allocation and task-splitting factor. The optimisation problem is solved by SCA-based algorithm. The results indicate that RSMA is reliable for communication in MEC under FBL constraints.
Version
Open Access
Date Issued
2025-02-25
Date Awarded
01/04/2025
License URL
Advisor
Clerckx, Bruno
Publisher Department
Department of Electrical and Electronic Engineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)