Green Task Offloading with Integration of Communication and Computation for LEO Satellite Computing Networks

Low Earth Orbit (LEO) satellite computing networks have opened up new possibilities for handling onboard tasks in space, while facing the challenges such as latency, energy consumption, and satellite battery lifespan degradation. In the existing studies, onboard computations are often fully offloaded to ground stations, resulting in large delays and energy inefficiencies. Moreover, the impact of satellite battery discharge depth on network lifespan has been neglected. To address these issues, this paper proposes a novel approach to offload tasks among satellites, which enables in-space processing and collaborative computation. To effectively extend the overall lifespan of the satellite network, we formulate an optimization problem which aims to maximize the weighted sum utility of energy efficiency and satellite battery lifespan degradation. Furthermore, a green task offloading strategy based on genetic algorithm is proposed to solve the problem. Simulation results demonstrate its benefits in reducing both satellites’energy consumption and discharge depth. Compared to the baseline, the proposed strategy prolongs the lifespan of the satellite network by 42.2%.