Adaptive network management for countering selective capture in wireless sensor networks

We propose and analyze adaptive network management for countering selective capture which aims to compromise critical sensor nodes close to the base station in a wireless sensor network (WSN) to block data delivery. We consider 3 countermeasures in the protocol design: (1) dynamic radio range adjustment; (2) multisource multipath routing for intrusion tolerance; and (3) voting-based intrusion detection. We identify the best protocol settings in terms of the best redundancy level used for multisource multipath routing, and the best number of voters and the intrusion invocation interval used for intrusion detection under which the lifetime of a WSN is maximized in the presence of selective capture which turns nodes into malicious nodes capable of performing packet dropping attacks and bad-mouthing attacks.