Blockchain-driven optimization of IoT in mobile edge computing environment with deep reinforcement learning and multi-criteria decision-making techniques

【Author】 Moghaddasi, Komeil; Masdari, Mohammad

【Source】CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS

【影响因子】2.303

【Abstract】The Internet of Things (IoT) presents complex challenges in task offloading, especially within Mobile Edge Computing (MEC) environments and under conditions of data insecurity. Addressing these challenges, particularly the balance between energy consumption, cost, and latency, necessitates intelligent decision-making strategies. This study introduces a blockchain-based offloading framework that leverages a Double Deep Q-Network (DDQN), a cutting-edge algorithm of Deep Reinforcement Learning (DRL) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), a Multi-Criteria Decision-Making (MCDM) technique. This approach strategically frames the MEC-based offloading problem using a Markov Decision Process (MDP), allowing the DDQN to learn the optimal policy. Subsequently, TOPSIS is applied to finalize offloading decisions based on predetermined criteria. The proposed framework significantly outperforms other strategies, demonstrating improvements in energy consumption, cost, and latency in the most complex scenarios by at least 35.83%, 4.65%, and 14.17%, respectively. These results underscore the efficiency and robustness of the presented multi-faceted approach in addressing the inherent complexities of task offloading within IoT environments.

【Keywords】Blockchain; IoT; MEC; Double deep Q-network; Multi-criteria decision-making

【发表时间】2023 2023 NOV 24

【收录时间】2023-12-14

【文献类型】实验仿真

【主题类别】

区块链技术-协同技术-物联网