Joint Offloading and Resource Allocation in Cooperative Blockchain-Enabled MEC System

Mobile edge computing (MEC) is regarded as an effective solution for processing computation-intensive tasks. Edge servers with limited resources which are deployed nearby mobile devices (MDs) suffer from many constraints than remote cloud, and relying on a single edge server have significantly reduced system performance. For relieving the mutual distrust and the conflict of interests problem between MDs and edge servers, and providing a data secure transmission MEC system. In this paper, we take MEC system security solved by blockchain technology into consideration, to utilize the cooperative computation offloading which enables edge servers to help each other in computation-intensive tasks. We investigate a long-term MEC system cost (energy consumption and processing delay) minimization problem while ensuring the transactional throughput rate of the blockchain network. Furthermore, we model the optimization problem as an Markov decision process (MDP). We design a dynamic offloading and resource allocation scheme based on Double Deep Q-network algorithm (DDQN). The simulation results show that our proposed algorithm is superior to others and it can improve system performance effectively.