Cost-Efficient Task Offloading for Satellite Edge Computing Systems
Xiumei Yang, Bo Hong · 2022 International Symposium on Networks, Computers and Communications (ISNCC) · 2022
Satellite edge computing is playing increasing important roles for low earth orbital (LEO) satellites to enable the onboard data processing. In this work, we investigate the satellite-terrestrial task offloading strategy for the satellite edge computing system where the task can either be processed onboard at the LEO satellite or remotely processed at the ground station. The offloading problem is formulated as a binary-integer programming problem in terms of the cost minimization for both the task delay and the energy consumption. A threshold-based solution is developed by using a novel measurement metric to evaluate the costs of different processing modes and further help determine the offloading decision. This method provides an improved energy-delay tradeoff performance with a low computational burden. Simulation results validate the effectiveness of the proposed method.