An Improved Neural based Cluster Heads Selection Mechanism for Wireless Sensor Network

Wireless Sensor Networks (WSNs) are used in every industry now because of the many benefits they provide. Over a thousand sensors are networked together to deliver higher-quality data tailored to each individual use case. But, for widespread area, selection of proper route for the data transmission is a biggest challenge for researchers. Clustering based routing is one of the most recent and better approach to transmit the data from source node to destination node via Cluster Head (CH). But, the selection of appropriate cluster head is biggest task and selection of CH is a critical aspect of WSNs. Selection of CHs, directly impacts the overall performance and efficiency of the network. An optimal CH selection mechanism should consider several factors, such as energy level, location, and communication capabilities of the sensor nodes, to ensure a balanced and energy-efficient distribution of responsibilities. So, in this research work, an improved neural-based CHs selection mechanism for WSNs is proposed. Here, the concept of Artificial Neural Network (ANN) is used to select CH efficiently during the data packets transmission with very less time span. This process directly impacts the efficiency of the network in terms of Energy Consumption, Packet Delivery Ratio (PDR), Delay, Alive Nodes and Throughput. The performance of the proposed CH selection mechanism evaluated in terms of such mentioned Quality of Service (QoS) parameters and compare with existing works. The simulation results of proposed architecture describe that neural-based CH selection mechanism provides better results as compared to traditional approaches.