Novel secure data aggregation schemes for wireless sensor networks

Abstract

Wireless sensor networks (WSNs) are capable of accumulating various physical phenomenon measures in the environment (like temperature, pressure, etc.) by using geographically distributed sensing devices. They help to monitor areas which are harmful and human inaccessible. The tiny nature of sensing devices with less storage, poor computing power and low battery resource makes the WSN starve from resource constraints. Research shows that the transmission consumes more energy compared to computation. Hence minimizing the communication cost becomes vital in energy starving WSN. The dense deployment of sensing devices leads to replication in the data being communicated and adds to the communication overhead. Data aggregation is one of the technologies which can remove the redundant data by aggregating them into a single data unit. The antagonistic and unsecured nature of the deployed regions makes the network prone to various attacks which affect the confidentiality, integrity and authenticity of the data communication. Most of the existing secure data aggregation schemes ensure confidentiality between the source node and base station (BS) by adopting privacy homomorphism (PH) based encryption. The integrity is verified in collective manner. But the presence of a single corrupted value makes this collective integrity verification to reject the whole received aggregate including both the valid and invalid data. This rejection of valid data consumes bandwidth of resource starving WSN. To address this problem, a secure data aggregation scheme (DIA-SSDAS) based on PH using slicing methodology is proposed. Novel child order based encoding mechanism for generating sliced messages is proposed which allows the BS to extract data slices smoothly from the aggregate and compute multiple statistical functions in a single query. newline