Q-learning-based semi-fixed clustering routing algorithm in WSNs

Abstract

In recent years, cluster-based routing protocols have emerged as a core technology for Wireless Sensor Networks (WSNs), attracting significant attention from researchers. This paper introduces a novel semi-fixed clustering algorithm, SFC-QL-IACO, designed to maintain energy balance in WSNs. The algorithm employs semi-fixed clustering to redistribute cluster nodes for initial load balancing and utilizes Q-Learning and enhanced ant colony optimization to construct data transmission paths. Clusters are dynamically adjusted when the energy difference exceeds a specified threshold to ensure energy balance. A dynamic energy threshold is implemented to prevent network disruptions caused by the depletion of cluster head energy, with cluster head rotation occurring as needed. Simulation results show that SFC-QL-IACO outperforms existing algorithms in terms of energy consumption, load balancing, and network lifetime.