The first rate-splitting multiple access experimental study for unicast, multicast and joint communication and sensing
File(s)
Author(s)
Lyu, Xinze
Type
Thesis
Abstract
In the recent Multiple Input Multiple Output (MIMO) system, numerous studies have demonstrated the theoretical and simulation-based advantages of Rate-Splitting Multiple Access (RSMA), showcasing its potential to enhance system performance across various scenarios. However, a gap remains in the form of experimental studies. These are essential to bridge the divide between theoretical models and practical implementations. This thesis bridges the research gap between the theoretical foundations and practical implementation of RSMA.
First, a prototype test-bed is developed to evaluate RSMA’s performance, comprising a two transmit-antenna base station and two single-antenna users. The experimental results demonstrate that RSMA outperforms Spatial Division Multiple Access (SDMA) and Non-Orthogonal Multiple Access (NOMA) in terms of both sum throughput and fairness performance. Notably, the advantages of RSMA become even more pronounced under practical conditions where the Channel State Information at Transmitter (CSIT) is quantized.
The second part of this thesis explores RSMA’s performance in Non-Orthogonal Unicast and Multicast (NOUM) scenarios. When compared to Multi-User Linear Precoding (MULP), RSMA demonstrates superior flexibility in managing and utilizing both unicast and multicast streams.This adaptability enables RSMA better to balance the demands of unicast and multicast communication.
The third part of this thesis investigates RSMA’s performance in overloaded scenarios. Specifically, a base station equipped with two transmit antennas serves two groups of users using multigroup multicasting. The results demonstrate that RSMA-based Multi-group Multicast (MGM) outperforms both NOMA-based MGM and SDMA-based MGM in terms of max-min fairness performance.
The final part of this thesis examines the application of RSMA in Integrated Sensing and Communication (ISAC) scenarios, specifically addressing the question of whether a dedicated sensing signal is necessary for ISAC. Both experimental and simulation results reveal that RSMA can achieve superior sensing performance without requiring a dedicated sensing signal, all while meeting the same throughput requirements.
First, a prototype test-bed is developed to evaluate RSMA’s performance, comprising a two transmit-antenna base station and two single-antenna users. The experimental results demonstrate that RSMA outperforms Spatial Division Multiple Access (SDMA) and Non-Orthogonal Multiple Access (NOMA) in terms of both sum throughput and fairness performance. Notably, the advantages of RSMA become even more pronounced under practical conditions where the Channel State Information at Transmitter (CSIT) is quantized.
The second part of this thesis explores RSMA’s performance in Non-Orthogonal Unicast and Multicast (NOUM) scenarios. When compared to Multi-User Linear Precoding (MULP), RSMA demonstrates superior flexibility in managing and utilizing both unicast and multicast streams.This adaptability enables RSMA better to balance the demands of unicast and multicast communication.
The third part of this thesis investigates RSMA’s performance in overloaded scenarios. Specifically, a base station equipped with two transmit antennas serves two groups of users using multigroup multicasting. The results demonstrate that RSMA-based Multi-group Multicast (MGM) outperforms both NOMA-based MGM and SDMA-based MGM in terms of max-min fairness performance.
The final part of this thesis examines the application of RSMA in Integrated Sensing and Communication (ISAC) scenarios, specifically addressing the question of whether a dedicated sensing signal is necessary for ISAC. Both experimental and simulation results reveal that RSMA can achieve superior sensing performance without requiring a dedicated sensing signal, all while meeting the same throughput requirements.
Version
Open Access
Date Issued
2024-12-17
Date Awarded
01/03/2025
License URL
Advisor
Clerckx, Bruno
Sponsor
UK Research and Innovation
Grant Number
EP/X52556X/1
EP/X040569/1
EP/Y037197/1
EP/X04047X/1
EP/Y037243/1
Publisher Department
Department of Electrical and Electronic Engineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)