Improvement of cluster head selection in leach protocol of wireless sensor network

Abstract

WSNs suffer from many issues such as coverage, security, energy-efficiency and localization. Among these issues, energy-efficiency is the most critical, as sensor nodes are battery operated, hence the need to optimize energy. Clustering technique has shown to be more suitable for energy efficiency, which is why LEACH protocol is considered. Despite that, the LEACH algorithm shows some drawbacks in the area of energy efficiency that needs to be enhanced in order to improve its performance. Therefore, since routing, communication and broadcasting between the nodes are done by the cluster heads, a modification in the LEACH algorithm was proposed where the improvement was done in cluster head selection to reduce the energy consumption. In this thesis, unlike the LEACH which uses the residual energy of the entire network, the cluster head selection takes into account the residual energy of each node to calculate the threshold value for next round. The cluster head selection uses a modified stochastic algorithm- the deterministic algorithm which calculates the residual energy of each node after each round to select the node with the highest residual energy as the cluster head. This approach makes the clustering algorithm adaptive to network dynamics as each node is considered to be a cluster head at a point. For simulation, the approach is implemented with LEACH in OMNET++ with Castalia and the results show that there is 3% increase in network lifetime.