摘要

The compressive sensing (CS) based data collection schemes can effectively reduce the transmission cost of wireless sensor networks (WSNs) by exploring the sparsity of compressible signals. Although many recent works explained CS as a symmetric cryptosystem, CS-based data collection schemes still face security threats, due to the complex deployment environment of WSNs. In this paper, we first propose two feasible attack models for specific applications. Then, we present a secure data collection scheme based on compressive sensing (SeDC), which enhances the data privacy by the asymmetric semi-homomorphic encryption scheme, and reduces the computation cost by sparse compressive matrix. More specifically, the asymmetric mechanism reduces the difficulty of secret key distribution and management. The homomorphic encryption allows the in-network aggregation in cipher domain, and thus enhances the security and achieves the network load balance. The sparse measurement matrix reduces both the computation cost and communication cost, which compensates the increasing cost caused by the homomorphic encryption. We also introduce a joint recovery model to improve the recovery accuracy. Experimental evaluation based on real data shows that the proposed scheme achieves a better performance compared with the most related works.