Using Bloom Filter to Generate a Physiological Signal-Based Key for Wireless Body Area Networks

Abstract

Wireless body area networks (WBANs) are often used to provide communication services for the data from the body. Since the data in WBANs are always closely related to individuals, they need to be kept secret with integrity. Key management is critical to data security. The resource-constraint biosensors make it difficult for traditional key management mechanisms to work well in WBAN. Using physiological signals to realize key agreement has the advantages of low overhead, timely key updating, and no key material predeployment and key store requirements, etc. The existing physiological signal-based key agreement schemes are always unable to balance the overhead and security well. To overcome these problems, we make two efforts. One is that we try to enhance the randomness of the interpulse-interval (IPI) from electrocardiograms in the process of digitizing physiological signals. And the other is that we attempt to use the Bloom filter rather than lots of chaff points to conceal the features exchanged for key agreement. The comparative analysis and experiments indicate that the proposed scheme can simultaneously achieve high security strength and low overhead.