Q-learning Aided and Elite Guidance Snow Ablation Optimizer for Heterogeneous Wireless Sensor Networks Coverage
摘要
To address the problem of maximizing sensor node coverage in heterogeneous wireless sensor networks (HWSNs), this study proposes a Q-learning aided and elite guidance snow ablation optimizer (SAO), named QEG-SAO. Our proposal has two novel components: (1) An adaptive diversity-driven step adjustment mechanism via Q-learning, which aims to overcome the limitation of fixed steps in SAO; (2) An elite guidance strategy is proposed to improve the quality of inferior individuals. QEG-SAO achieves superior performance compared to SAO, its variants, advanced meta-heuristic algorithms, and entries from the CEC competition, as validated on the CEC2017 benchmark and simulations of two-dimensional sensor node coverage considering a variety of maximum sensor radii and error rates. Notably, it achieves 4.07% and 9.43% higher coverage than SAO in our defined scenarios.