Research on High Confidence Resource Allocation Technology in Edge Computing

Abstract

This paper studies resource allocation and security performance. Aiming at the problem of privacy disclosure caused by eavesdropping during task unloading in single cell and multi user scenarios, the physical layer security technology is used to formulate corresponding confidentiality measures. Facing the dynamic change of the wireless channel state of the system and the computing power required by users, taking reducing the average processing delay of tasks as the optimization goal, the multi-user task partial migration problem is modeled as a joint optimization problem of computing resources and power resources under the constraints of security set-tings and energy constraints, and the system is established as a Markov decision process model, and proposes a resource allocation algorithm based on physical layer security and depth deterministic policy gradient. Simulation results show that the algorithm can approach the optimal performance, has adaptability and lower computational complexity, can make better unloading strategy and resource allocation scheme in trusted environment, effectively reduce the average task processing delay and improve the robustness of the system.