Adaptive Video Streaming in Integrated Satellite-Terrestrial Networks: A Low-Complexity Model Predictive Control Approach

Abstract

Adaptive video streaming technology is regarded as a powerful method for enhancing the quality of experience (QoE) for video services. However, its effectiveness is hindered by the highly fluctuating and limited coverage nature of terrestrial networks. Recent advancements in Integrated Satellite-Terrestrial Networks (ISTN) offer a promising solution by bolstering the reliability and coverage of terrestrial networks. In this letter, we investigate a video streaming system within an ISTN, utilizing joint bitrate and channel adaptation. Our primary objective is to develop an algorithm that optimizes both bitrate and channel selection to maximize streaming QoE. To accommodate the unique characteristics of terrestrial and satellite channels, we propose a satellite channel planning module and a terrestrial channel sensing module to forecast future bandwidths for each. Leveraging these predictions, we employ a model predictive control approach for sequential decision-making, formulating the MPC problem as a mixed-integer non-linear programming task. To mitigate the computational burden associated with solving MINLP, we introduce a dual-stage iterative optimization method to achieve efficient decision-making. Experimental results demonstrate that compared with the most competitive benchmark, the proposed method enhances QoE by 11% while reducing decision-making overhead.