Fault Containment based Self Stabilized Fuzzy Relevance Clustering Algorithm

Abstract

In extensive scale wireless ad-hoc sensor networks for the support of self-organization and enable routing, clusters have been introduced. Robustness is a key issue due to the dynamic environment in these kinds of wireless networks. Even though numerous algorithms have already been designed for clustering, only a few of the algorithms addressed the issue of fault containment feature along with the self-stabilization in sensor networks. A system delivers fault-containment if the probability that an application fails is related to the number of resources used by that application, not to the total number of resources in the system. Thus, theproposed algorithm prevents increasingerrorsin the network at small-scale fault incidencecondition. It allowseveryCH to interact its neighbour CHs to build time concurrently and energyefficient routes. The clustering approach during cluster formation constructs spanning trees among CHs and Cluster Members (CM) using a self-stabilizing approach to regulateappropriate routes for clusters. Decreasing the amount of unusable cluster-heads in the caused topology is the main objective of implement rules for the nodes. The experimental result for the proposed approach is carried out using NS2 simulator where it is shown that performance metrics of the proposed approach is well when contrasted with the current self-stabilized, Fault containment and traditional FRCA clustering algorithm.