A hybrid Kalman-RBF model for optimized cluster-based routing in wireless sensor networks

Wireless Sensor Networks (WSNs) play a major role in real-time monitoring and data transmission, yet the existing network clustering and routing protocols face significant challenges in terms of unbalanced cluster formation, delay and high latency. This research work proposes an optimized WSN routing algorithm, the Multi-Dimensional Kalman and Radial Basis Function (MDRBF) method by integrating dynamic cluster head selection based on node distance ranking and residual energy. Further, Kalman filtering is used for noise reduction and data aggregation and Radial Basis Function for adaptive routing optimization. The performance of MDRBF is evaluated using NS 3.26 simulation, considering the key performance metrics such as End-to-End delay, Packet Delivery Ratio (PDR), throughput, routing overhead and residual energy. The experimental results of proposed MDRBF model are then compared with other existing protocols like LEACH, POHDA, CDAS, DICA and ESEERP. The results show that MDRBF significantly reduces delay to 8 ms, increases PDR to 93%, improves node residual energy and reduces routing overhead with optimized energy consumption when compared to other algorithms.