Neural based Energy-Efficient Stable Clustering for Multilevel Heterogeneous WSNs

This paper presents a Neural based Energy-Efficient Stable Clustering (NESC) for Multilevel Heterogeneous wireless sensor networks (MHWSNs). In NESC protocol, the sensor nodes are selected as cluster heads (CHs) by employing the multi-layer back propagation model of neural network. The training of neurons is done on the basis of normalized energy and distance factors which helps in the selection of proper CHs which increases the network lifetime, throughput and reliability. Simulation results justified that NESC protocol achieves better network performance in terms of network lifetime, energy consumption, and throughput than existing routing protocols (i.e., LEACH, SEP, DEEC, and EDCS) for MHWSNs.