Optimal resource allocation for satellite‐aided collaborative computing among multiple user pairs

Abstract

SummarySatellite‐aided communication networks play a vital role for users in remote areas. With the increasing enhancement of users' computing capacities for satellite‐aided communication networks, a part of the data can be compressed first at the source node and then decompressed at the destination node to relieve the burden of the transmission. In this way, collaborative computing among multiple user pairs is achieved. Due to the limitations of the computing and communication resources, how to realize collaborative computing efficiently and allocate the communication resource optimally is the problem that needs to be addressed. In this paper, a collaborative computing and resources allocation (CCRA) method is proposed to minimize the weighted‐sum latency of multiple user pairs formed by two nodes with computing capacities that exchange data via the satellite. With CCRA, the upper bound of the end‐to‐end latency is defined and used to decompose the problem of joint computing and resource allocation for multiple user pairs into independent problems for every user pair. And then, the optimal solutions for collaborative computing and resource allocation can be achieved with closed‐form solutions and low complexity. Numerical results verify the effectiveness of CCRA in latency optimization.