DPRL: Task Offloading Strategy Based on Differential Privacy and Reinforcement Learning in Edge Computing

Abstract

Mobile edge computing has been widely used in various IoT devices due to its excellent computing power and good interaction speed. Task offloading is the core of mobile edge computing. However, most of the existing task offloading strategies only focus on improving the unilateral performance of MEC, such as security, delay, and overhead. Therefore, focus on the security, delay and overhead of MEC, we propose a task offloading strategy based on differential privacy and reinforcement learning. This strategy optimizes the overhead required for the task offloading process while protecting user privacy. Specifically, before task offloading, differential privacy is used to interfere with the user's location information to avoid malicious edge servers from stealing user privacy. Then, on the basis of ensuring user privacy and security, combined with the resource environment of the MEC network, reinforcement learning is used to select appropriate edge servers for task offloading. Simulation results show that our scheme improves the performance of MEC in many aspects, especially in security and resource consumption. Compared with the typical privacy protection scheme, the security is improved by 7%, and the resource consumption is reduced by 9% compared with the typical task offloading strategy.

Safe-Learning-Based Location-Privacy-Preserved Task Offloading in Mobile Edge Computing

“…In recent years, many works have utilized differential-privacy-based location perturbation techniques to effectively protect users' location privacy [14,18]. Differential privacy technology has superior privacy protection effects and can prevent attackers from re-identifying data based on known background knowledge [18,25].…”

“…In [29], Liu et al propose a privacy-preserving distributed deep deterministic policy gradient (P2D3PG) algorithm that solves the problem of maximizing the distributed edge caching (EC) hit rate under privacy protection constraints in a wireless communication system with MEC. In [14], Zhang et al studied differential privacy and RL task transfer strategy, established an MEC system model, and designed a four-layer policy network as an RL agent, but lacked a balance between privacy and computation offloading performance. To solve the above problems, our work proposes an RL-based algorithm that achieves a balance between privacy protection and computation offloading performance by combining differential privacy and RL technology.…”

“…The system lacks a dynamic balance between the computational cost of MEC systems and users' privacy requirements. Notably, the protection of user location privacy can lead to an increase in the computational burden on MEC servers [14].…”