K-means based LEACH protocol Implementation

WSN, Wireless Sensor Networks comprise a group of hardware sensors which are arranged in a wide area. The Sensor nodes are usually placed in areas where human access is not so easy. The sensors are normally powered by batteries. We should either replace the node or recharge it when the battery is empty. Thus, one of the biggest challenges in wireless sensor networks is to save energy and extend battery life. There are several energy efficient mechanisms that exist. Choosing an appropriate routing strategy is one of the best solutions to save energy. One of the best routing strategies is the Low Energy Adaptive Clustering Hierarchy Algorithm (LEACH) to reduce energy consumption between nodes. This paper explains the normal LEACH protocol and a LEACH protocol based on K-means clustering. In the second case, clustering is performed using the machine learning algorithm K-means. Instead of selecting a random cluster head, this algorithm selects the cluster head by computing the residual energy of each node and the Euclidean distance between cluster heads and other nodes. Efficient clustering and cluster head selection increase the number of transactions. Here, a comparison between the normal LEACH and the K-means-based LEACH algorithm is explained. According to the comparison, the second method increases the performance of the traditional LEACH algorithm.