Dynamic clustering and analysis for sensor networks

The most important challenge in wireless sensor networks (WSNs) is to improve the operational efficiency in highly resource constrained environment based on dynamic and unpredictable behaviour of network parameters and applications requirement. In this paper we have proposed a method for clustering and their analysis to study the cluster formation, their behaviour with respect to the system parameters and applications requirement. The technique involves the adoption of computational intelligence to form clustering. We have used Kohonen self organization map neural network (KSOM-NN) for the above computations. The simulations are carried out to evaluate the performance of the proposed method with respect to different parameters of sensor networks and applications requirement.