Fuzzy logic based compromised node detection and revocation in clustered wireless sensor networks

Wireless sensor networks contain a large number of nodes which are unattended in nature where an adversary can physically capture and compromise the nodes and later inject a range of attacks with these compromised nodes. In order to minimize the damage caused by the compromised nodes, the system should identify and remove them as early as possible. Researchers recently have proposed a number of compromise detection schemes in sensor networks. Trust based schemes are capable of detecting compromised nodes but do not revoke them because of the overhead of false positives. Other methods like software attestation have got overhead in the periodical node attestation. We propose a cluster based node compromise node detection and revocation scheme which reduces the limitations of the existing schemes. In this scheme the concept of Fuzzy logic is used to make a decision over the clusters whether they contain suspicious nodes. After identifying the suspected clusters the software attestation is performed against the nodes which lead to the identification and revocation of compromised nodes from the network. Our proposed scheme shows that it performs well in the presence of false positives and false negatives.