A Novel Load Balancing Aware Graph Theory Based Node Deployment in Wireless Sensor Networks

Wireless sensor networks are one of the drastically growing networks in recent times. It supports a greater number of applications in real time industries and automation sectors. All the applications of WSN comprises of numerous numbers of sensor nodes where they are deployed in a manner according to the application needs. Sensor nodes sense, monitor, record, receive and transmit any kind of data based on its manufacturing motivation. Sensors are cheap in cost, small in size and restricted in energy efficiency. Owing to their remote deployment and small size, they are provided with limited battery power. When the energy drains out, it becomes dead and thus causes a bottleneck in the communication process. Subsequent failure of nodes due to inefficient routing methods tends to degrade the network lifetime and overall performance. Several earlier research methods were used for improving the network lifetime and reliability of node’s connection and communication. Most of the methods were not able to provide optimal performance towards enhancing overall QoS which happens to be a collective attribute. This paper proposes a novel Graph Theory Clustering method (GTC) to do node clustering, data aggregation and load balancing in WSN. The entire process of GTC has three modules or stages. First stage focuses on clustering and cluster head selection. Second stage focuses on distance computation. The third stage focuses on shortest path calculation for finding shortest route to do data transmission where it reduces the energy level, delay, packet loss, and load balancing. The proposed GTC is simulated in NS2 software, and the performance compared against existing benchmark methods. Various metrics have been computed and analyzed and superior performance of proposed GTC is observed in each case.