Dynamic Resource Management for Enhanced QoS in Collaborative Edge-Edge Industrial Environments

In this paper, we focus on dynamic resource manage-ment within the heterogeneous Collaborative Edge-Edge Indus-trial Environments (CEIE) architecture to better satisfy diverse industrial applications, formulating a multi-stage Mixed Integer Nonlinear Programming (MINLP) problem to minimize system costs. To reduce the computational complexity of solving the MINLP, we decompose the original problem into multi-user task offloading, Communication Resource Allocation (CmRA), and Computational Resource Allocation (CpRA) problems. To tackle the complexities of state and action spaces, we introduce an improved Double Deep Q-Network (iDDQN) strategy, enhanced with a priority experience replay mechanism. This innovation prioritizes the replay of more critical experiences frequently, thereby expediting the convergence of the Q-value function and reducing the training duration. Extensive simulations show that our proposed optimization scheme can reduce the average weighted system cost by up to 49.39%. Moreover, testbed ex-periments demonstrate that the proposed algorithm can reduce the average system cost by as much as 50.65% in real-world applications, outperforming existing methods.