Secure Data Aggregation Solution Based on Dynamic Multiple Cluster Key Management Model

Abstract

A dynamic multiple cluster key management model based on identity is designed in wireless sensor network. The key management in the model is cluster-based. The model can realize the identity authentication independent of the third party, thus no use for infrastructure to sustain the key distribution center. The model consumes little storage space and can resist the collusion attack because every sensor node in the model is just required to store its own private key and the public key factor matrix of the cluster that it belongs to. A secure data aggregation solution based on the model is proposed. The communication process is described by formal methods, and the security of the proposed solution is validated using BAN logic. The energy consumption during the handshaking process is discussed, which indicates that it is feasible to use the identity-based cryptography in the wireless sensor networks. The simulation shows that, compared with two other typical data aggregation schemes TAG and CPDA, the proposed solution has preferable security, reasonable communication consumption and desirable success ratio.