| Summary: | 博士 === 國立中興大學 === 資訊科學與工程學系所 === 97 === Over the last two decades, we have seen growing importance placed on research in wireless sensor networks (WSNs). A WSN is composed of hundreds of thousands tiny devices, some headers, and some sink nodes. The headers and forward nodes are more powerful than sensor nodes. A sink node is usually a server and is located at a safe location. Generally, there are four main constraints of a sensor node, namely battery energy, computation capability, memory size, and communication protocols. Many literatures investigate security issues including: key management, location deployment, concealed data aggregation (CDA), routing paths and so on. However, the issue of saving power is aimed in this dissertation. To deal with this problem, two rekeying schemes and a concealed data aggregation scheme under power-saving are proposed. This dissertation is composed of two parts, one is rekeying schemes based on code slices, and another is concealed data aggregation scheme based on secret sharing.
In the first part, two reliable and efficient rekeying schemes for wireless sensor networks are proposed. One is an efficient rekeying scheme based on code slices, and another is an efficient key management scheme based on genetic algorithms. The sensor system is divided into sinks, headers and sensor nodes. The sink nodes find out a lower power consumption rekeying function and broadcast it randomly. The headers and sensor nodes will collaborate and assemble the code slices to finish the rekeying. Experiments show that the proposed schemes can find the lower power consumption rekeying functions and consume the energy smoothly.
In end-to-end encryption, how to maintain security has become a challenging research issue. To save the overall energy resources and maintain the security, computation cost needs to reduce as well as the amount of encrypted data transmitting through the wireless sensor networks. One plausible approach is to consolidate the encrypted data along the routing path. In the second part, a novel concealed data aggregation scheme is proposed to save power by reducing the delivering package sizes. It is provably secure up to some fixed number of compromised nodes.
|
|---|
