Privacy-Preserved Computation Offloading Scheme in 5G enabled IoT Based on Smart blockchain

Abstract

To effectively utilize multiple 5G resources (such as computing, communication, storage and service resources), provide low delay offload decisions, and meet data privacy, we propose a framework to integrate blockchain, federated learning (FL), and mobile-edge computing (MEC) into the Internet of Things (IoT) system. However, the insufficient throughput of the blockchain may hinder the efficiency of the overall system. At the same time, due to the dynamic characteristics of IoT and MEC systems, the whole framework involves high-dimensional features and large-scale actions, and the optimization problem is very complex and challenging. Therefore, we design the whole system flow process as a Markov decision process (MDP) sequence by defining state space, action space, and reward function. The deep deterministic policy gradient (DDPG) algorithm is adopted to dynamically select and adjust action to ensure the maximum long-term return of the system. Finally, the simulation results show that our scheme can significantly improve the performance of system.