A GA-BASED KEY-MANAGEMENT SCHEME IN HIERARCHICAL WIRELESS SENSOR NETWORKS

Abstract

In this paper, we propose a novel key management scheme based on genetic algorithms to fulfill the requirement of power control, memory usage, and computation security in a hierarchical wireless sensor network. We divide our scheme into three parts: the sink node, the header nodes and the sensor nodes. The sink node first uses genetic algorithms to generate appropriate key-generating functions, and then deliver them to header and sensor nodes. Each possible key-generating function is encoded as a chromosome, which is feasible if it satisfies the specified power-consumption constraint. Its fitness is set as the entropy measure for evaluating key distribution. The final several key-generating functions are then gathered for rekeying. The headers and the sensor nodes can then assemble common keys from the keygenerating functions for communication. The proposed scheme is simple, efficient and secure if the sensor nodes cannot be compromised within a threshold time bound. Experiments are also made to show the performance of the proposed approach.