Reputation based clustering algorithm for security management in ad hoc networks with liars

Clustering in ad hoc networks consists in dividing the network into clusters (groups) managed by elected nodes called clusterheads. This technique has been used for different goals as routing efficiency, transmission management and information collection. As far as we know, no existing clustering algorithms have taken into account the existence of malicious nodes for clusterheads election and maintenance. These nodes can lie to be elected as clusterheads. Consequently the network might be managed by most of them. To solve this problem, we propose a reputation based clustering algorithm (RECA) that aims to elect trustworthy, stable and high energy clusterheads that can be used to manage the security of the network. Simulations were conducted to evaluate RECA performances in the presence of liars. Results show that it converges to a stable and convenient network division into clusters with no untrustworthy clusterheads and mainly one hop members.