Decreasing Big Data Application Latency in Satellite Link by Caching and Peer Selection

Although satellite communications have attracted wide attention in 6 G integrated space-air-ground networks, the link delay between the satellite and the satellite base station is a major issue with the advent of the era of big data. Considering this problem, this paper proposes cooperative strategies between peers in order to decrease the transmission latency for both downlink and uplink, respectively. In the downlink, we establish social relationship to find peer nodes who can cache contents and design a caching placement algorithm based on greedy methods to configure the cache. The approximation ratio of the solution obtained by the cache placement algorithm is no less than
$1-1/e$
of the optimal solution. In the uplink, when channel conditions are severe, some users can be selected as relays to ensure the quality of communication. Such that another social relationship is established to prevent the disclosure of privacy in the process of data forwarding. We establish a dynamic optimization model, and propose a peer selection algorithm. Simulation results show that the performance of cache placement algorithm is more efficient other algorithms in the aspect of average latency and hit ratio in the downlink, and the peer selection algorithm outperforms other algorithms in terms of average channel rate and handover ratio in the uplink.