PSO-Based K-means Algorithm for Clustering Routing in 5G WSN Networks

With the development of the Internet of Things, Wireless Sensor Network (WSN) in 5G networks is becoming more and more important in the field of information and communication technology, and is widely used in many scenarios. However, WSN usually has limited energy due to its compact structure. For energy consumption issues, the hierarchical routing architecture has been considered that is an extremely effective method to save network energy, but uneven network clustering and unreasonable Cluster Head (CH) will lead to unbalanced network energy consumption and reduce the network lifetime. In this paper, we intend to use the K-means algorithm for network clustering. Considering K-means algorithm is sensitive to the Initial Center (IC) and is easy to fall into the local optimum, we use Particle Swarm Optimization algorithm (PSO) to optimize the initial clustering center of K-means to obtain the optimum clustering. After the network clustering is completed, we comprehensively considers the Sensor Node’s (SN) energy and SN’s location factors for CH selection, and dynamically updates the weight of the factor according to the remaining energy of the SNs. Simulation results show that the proposed protocol performs well in balancing the network’s energy consumption and extending lifetime of the network.