An adaptive hybrid differential Grey Wolf Optimization algorithm for WSN coverage

To address the issues of large blind spots and uneven distribution in traditional Wireless Sensor Network (WSN) node deployment, we propose an Adaptive Hybrid Differential Grey Wolf Optimization (AHDGWO) algorithm for solving the WSN coverage problem in 2D area. Firstly, an adaptive exponential convergence factor is designed, allowing each individual to adjust global exploration and local exploitation adaptively. Secondly, by integrating the concept of differential mutation, an hourglass-shaped random search area is established. This not only prevents blind search but also bolsters the algorithm's global exploration capabilities. Then, on the CEC2022 test set, the mean error and standard deviation of the AHDGWO algorithm are compared with those of seven algorithms, including the standard GWO, excellent evolutionary algorithms from recent years, and advanced variants of GWO. The convergence curve graphs of the AHDGWO algorithm demonstrate superior accuracy and convergence speed. Meanwhile, results from two statistical tests and box plots also indicate that the AHDGWO algorithm possesses significant advantages and excellent stability. Finally, through simulation experiments conducted on two scales, the performance of the AHDGWO algorithm is evaluated in 2D WSN scenarios with varying numbers of sensors. The experimental results show that the coverage achieved by the AHDGWO algorithm after optimization surpasses that of the other seven compared algorithms, while ensuring network connectivity. Specifically, when the number of sensors is 50 and the coverage area is 10,000 square meters, the average coverage rate can reach 98.71%, indicating that the algorithm exhibits good practicality and scalability in addressing WSN coverage issues.