When Multipath QUIC Meets Model Predictive Control and Band Sparse Network Coding: A Novel Multipathing Solution for Video Streaming Over Heterogeneous Wireless Networks
Year: 2025
Type: article
Abstract: Multipath Quick UDP Internet Connections (MPQUIC) integrated with network coding offers a promising approach to improving the Quality of Experience (QoE) for video services over heterogeneous wireless networks. However, a significant challenge arises when encoding nodes transmit potentially redundant packets while awaiting decoding acknowledgments (ACKs) from endpoints. This behavior can limit effective transmission rates, thereby degrading real-time streaming performance and user QoE. In this paper, we propose MP2-QUIC, which addresses these challenges through a novel adaptive Model Predictive Control (MPC) framework for MPQUIC that optimizes both congestion window and encoding redundancy parameters via a discrete state transition model. By incorporating operating point linearization and leveraging the Central Limit Theorem, MP2-QUIC effectively enhances the control performance and effective throughput of the model in heterogeneous wireless network environments. MP2-QUIC further employs Band-Sparse Network Coding (Band-SNC) to minimize computational complexity at endpoints, while utilizing queuing theory principles to determine optimal encoded packet quantities. This integrated approach significantly enhances end-user QoE, and the experimental results demonstrate MP2-QUIC’s superior performance compared to existing MPQUIC encoding solutions, yielding a 68.85% reduction in peak decoding overhead and marked improvements in Peak Signal-to-Noise Ratio (PSNR).
Authors: Yuanlong Cao, Haopeng Zhang, Ming Jiang, Yirui Jiang, Jinquan Nie
Institutions: Jiangxi Normal University, Cranfield University
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: G166464655, G7203076342
OpenAlex ID: W4411337107