Security Topology Control Method for Wireless Sensor Networks with Node-Failure Tolerance Based on Self-Regeneration

Abstract

Wireless sensor networks are often deployed in hostile and unattended environments. The nodes will be failure by fault, intrusion, and the battery exhaustion. Node-failure tolerance is an acceptable method to improve the networks' lifetime. Then two key problems for topology control are presented: first, how to get a node-failure topology when there is intrusion from the nodes of hostile enemies? second, how to sustain this node-failure topology with all deployed nodes being exhausted ultimately? We propose a novel approach for topology control and prove that it is node-failure tolerant. The approach contains three phases: topology discovery, topology update, and topology regeneration. A tricolor-based method is proposed to build architecture with high tolerance ability and some security protocols are employed to preclude the hostile nodes in discovery phase. In update and regeneration phases, the newly deployed nodes are regarded as renewable resource to fill in the consumed energy, enhance the debased node-failure tolerance ability, and prolong network lifetime, and a security protocol with forward and backward secrecy is devised to adapt the topology changed by node failure and node joining. Some attributes of the presented method are shown by simulations, and differences are given by comparison with related work.