Novel Optimization Framework for Energy-Efficiency-Based Resource Allocation and Multihop Offloading in Blockchain-Enhanced IoT

The integration of blockchain with Mobile Edge Computing (MEC) enhances Internet of Things (IoT) security but increases energy consumption. Traditional single-hop/fixed-relay offloading schemes significantly limit coverage and energy efficiency. To address these issues, this paper proposes an innovative optimization framework combining multi-hop offloading with blockchain-enhanced IoT. First, a Modular Cluster Head IoT Device (MCHID) is designed to reduce energy consumption through multi-hop offloading using only the active Local Wireless Module (LWM), while precisely modeling blockchain consensus protocol computations. Second, a parameterized action-space Markov Decision Process (MDP) is developed to jointly optimize device state switching, resource allocation, and multi-hop routing. Finally, a Hybrid Trust Region Policy Optimization (HTRPO) algorithm addresses hybrid-action-space limitations in traditional reinforcement learning, effectively reducing energy consumption and overhead. Experimental results demonstrate both the effectiveness and superiority of the proposed approach.