On QoE Optimization for Live Streaming Services in Multi-User Uplink RSMA Networks = 다중 사용자 업링크 RSMA 네트워크에서 라이브 스트리밍 서비스의 QoE 최적화

다국어 초록 (Multilingual Abstract)

The rapid proliferation of IoT devices and the increasing demand for live streaming platforms have exacerbated significant network congestion, degrading the quality of experience (QoE) of uplink streamers. The exponential growth of high-quality real-time video streaming services over cellular networks, particularly in heterogeneous environments facilitated by 5G networks, has underscored the need for reliable wireless communication via uplink network services. To address this challenge, Rate Splitting Multiple Access (RSMA) has emerged as a promising interference management scheme in multi-user live communication systems. This paper explores an RSMA-enabled BARMAS optimization method for uplink communication systems to enhance network performance and ensure superior video streaming quality. We formulate an optimization problem that simultaneously considers the popularity and retention rate of live video streams to maximize the video bitrate for average streamers. This complex problem is modeled as a Markov Decision Process (MDP) and subsequently addressed using a reinforcement learning framework, specifically the Deep Deterministic Policy Gradient (DDPG) technique. The simulation results demonstrate that the proposed DDPG-BARMAS method significantly outperforms existing uplink communication algorithms, highlighting its potential as a robust solution for future wireless uplink live streaming services. Keywords: Live Video Streaming Services,Multiuser Uplink RSMA, DeepDeterministic Policy Gradient (DDPG), Optimization of the Bitrate, Superior Uplink VideoQuality.

목차 (Table of Contents)

• Abstract i
• List of Figures v
• List of Tables vi
• List of Abbreviations vii
• 1 Introduction 1
• 1.1 Motivations 2
• 1.2 Contributions 2
• 2 RelatedWork 5
• 2.1 Optimizing Decoding Order in Live Video Streaming via RSMA Technique: 5
• 2.2 Maximizing Average Video Bitrate for Streamers by Leveraging the Deep
• Deterministic Policy Gradient (DDPG) Algorithm 7
• 3 ProblemStatement 11
• 3.1 Channel Model 11
• 3.2 Service Model 14
• 3.3 Optimisation Problem 16
• 4 Proposed Solution 17
• 4.1 Continuous Decoding Order for uplink RSMA 17
• 4.2 Optimization of Uplink RSMA Systems usingMarkov Decision Processes . 18
• ii
• 4.3 State Space 19
• 4.4 Action Space 20
• 4.5 Transition Probability 21
• 4.6 Reword Function 22
• 4.7 Optimized DDPG Framework for BARMAS Control Systems 25
• 4.8 Effective Training and Validation Strategies for Algorithm 28
• 4.9 Optimal Strategies for Computational Complexity Analysis 31
• 5 Simulation Results 35
• 5.1 Evaluation of Performance and Analytical Discussion 35
• 5.2 Simulation Configuration and Environment Setup 35
• 5.3 Analysis of Streamer Popularity 36
• 5.4 Evaluating Retention Rates for Streamers 39
• 5.5 The Uplink RSMANetwork Architecture 42
• 5.6 Uplink Network Live Streaming Service Architecture 43
• 5.7 Analysis of Convergence Dynamics 46
• 5.8 Optimized Selection of Uplink Decoding Order 51
• 5.9 Comparative Analysis of Bitrate Fluctuations Across Multiple Streamers 52
• 5.10 Comprehensive Performance Analysis 53
• 6 Comparative Analysis of Bitrate andQoEPerformance in Live Video Streaming
• Algorithms 60
• 7 Concluding Remarks 62
• iii
• References 63
• 국문초록 69
• Acknowledgements 70
• iv