A Two-Layer Caching Model for Content Delivery Services in Satellite-Terrestrial Networks

Abstract

With the development of satellite communication technologies and user requirements for pervasive network access, there is a trend to integrate satellites into the terrestrial network infrastructure. Such kind of satellite-terrestrial network is often used for content delivery services as satellites are with wide-area coverage. In terrestrial networks such as Internet, in-network caching has been proved to be an effective method to improve the network performance in terms of throughput and delay. Based on this observation, we involve caches in the satellite-terrestrial networks. Particularly, a two-layer caching model is proposed for content delivery, where caches placed in the ground stations constitute the first caching layer and caches deployed in the satellite forms the second one. On the satellite, to make full use of the broadcast advantage, we set a window to aggregate the requests for the same files from ground stations. These requests will be served by one satellite broadcasting when the aggregation window expires. Our goal is to minimize the downlink and uplink satellite bandwidth consumption, which requires joint caching optimization between the satellite and ground stations. We formulate the joint caching optimization problem as a nonlinear integer programming problem. Furthermore, a caching strategy based on the genetic algorithm is proposed to solve the problem efficiently. The simulation results show that the proposed caching strategy significantly outperforms content popularity based and random caching strategies in terms of satellite bandwidth consumption.