Edge-Assisted Satellite Ground Network Task Offloading and Resource Allocation Strategy

Emerging as an innovative architectural paradigm for networks, satellite-ground network is the key to the future Internet of Everything. Focusing on the challenge of computing offloading of ground users in the edge computing scenario of low earth orbit (LEO) satellites, an edge-assisted satellite ground network task offloading and resource allocation strategy is proposed. Firstly, the communication model and computing model between users and base stations and LEO are analyzed. Secondly, a system energy consumption minimization problem is established by jointly considering the maximum delay constraint, task offloading, communication between base stations and LEO, and computing resource allocation. Finally, the alternating optimization algorithm is utilized to split original non-convex formulation into a pair of sub-problems: task offloading and resource allocation. The non-cooperative game algorithm based on sub-gradients is used to solve the optimal task offloading strategy, and the Lagrange multiplier method is used to optimize communication and computing resource allocation problems. As illustrated by the simulation, the proposed joint optimization algorithm has good convergence, and compared with other algorithms, the system energy consumption is reduced by at least 10.61 %, and the average service delay is reduced by at least 12.41 %, which has good performance.