A Computation Offloading Strategy in Satellite Terrestrial Networks with Double Edge Computing

Abstract

Mobile edge computing (MEC) can efficiently minimize computational latency, reduce response time, and improve quality-of-service (QoS) by offloading tasks in the access network. Although lots of edge computation offloading schemes have been proposed in terrestrial networks, the hybrid satellite terrestrial communication, as an emerging trend for the next generation communication, has not taken edge computing into consideration. In this paper, a novel satellite terrestrial network with double edge computing is introduced to reap the benefits of providing computing service for remote areas. A strategy is designed to solve the problem of efficiently scheduling the edge servers distributed in the satellite terrestrial networks to provide more powerful edge computing services. For the purpose of allocating satellite edge computing resource efficiently in the strategy, a double edge computation offloading algorithm is proposed to optimize energy consumption and reduce latency by assigning tasks to edge servers with minimal cost. Numerical results verify that the proposed algorithm can reduce the average latency and the energy consumption by approximately 45% and 49% respectively. The study on the number selection of utilized satellite edge servers provides an insight for following studies of edge servers scheduling in satellite terrestrial networks.