User Satisfaction-Oriented Video Streaming in Satellite Terrestrial Integrated Networks

In this paper, we investigate video streaming for satellite broadcasting applications in satellite terrestrial integrated networks, employing the low complexity enhancement video coding scheme. Considering the inherent limitations of satellite broadcasting, we design a collaborative video streaming method where the base layer (BL) is transmitted via satellite links, while terrestrial links are utilized for transmitting the enhancement layer (EL). This design seamlessly integrates with existing satellite broadcasting schemes since the BL can utilize any standard video codec, while the EL significantly enhances video quality when received via terrestrial links. To accommodate dynamic network conditions, we formulate a video segment delivery problem aimed at maximizing the overall user satisfaction by determining the optimal number of video segments with ELs downloaded for each broadcasting channel. To solve the problem, we propose a group knapsack-based EL segment downloading (GKELD) algorithm, which transforms the user satisfaction maximization problem into a group knapsack problem and then solves it by devising a dynamic programming-based approach. Simulation results demonstrate that our proposed algorithm significantly outperforms benchmark methods, achieving higher overall audience satisfaction under limited network resources.

I. Introduction

The satellite terrestrial integrated network (STIN) has garnered significant attention from both academia and industry due to its capability to utilize complementary advantages of terrestrial and satellite network segments [1], [2], [3]. The STIN is promising in supporting various applications due to high-speed and low-latency terrestrial transmissions, and the seamless coverage and highly resilient satellite communications. Specifically, video delivery is crucial as it plays an important role in applications such as multimedia infotainment, immersive virtual/augmented reality, and video surveillance.