Secure Computation Offloading in Blockchain based IoT Networks with Deep Reinforcement Learning

For current and future Internet of Things (IoT) networks, mobile edge-cloud offloading (MECCO) has been regarded as a promising means to delay-sensitive IoT applications. However, offloading mobile tasks to cloud is vulnerable to security issues due to malicious mobile devices(MDs). How to implement offloading to alleviate computation burdens at MDs guaranteeing high security in mobile edge cloud is a challenging problem. this paper, we investigate simultaneously the security and computation problems in a multi-user MECCO system with blockchain. First, to the offloading security, we propose a trustworthy access control using, which can protect cloud resources against illegal offloading. Then, to tackle the computation management of authorized MDs, we a computation offloading problem by jointly optimizing the offloading, the allocation of computing resource and radio bandwidth, and smart usage. This optimization problem aims to minimize the long-term system of latency, energy consumption and smart contract fee among all MDs. To the proposed offloading problem, we develop an advanced deep learning algorithm using a double-dueling Q-network. Evaluation from real experiments and numerical simulations demonstrate the advantages of our scheme over existing approaches.