Research On Secure Data Aggregation In Wireless Sensor Networks

Wireless sensor network (WSN) is an ad-hoc network which is composed of a collection ofresource constrained sensor nodes that deployed in the monitoring area. Data aggregation is acritical technique of wireless sensor networks. It is used for removing the redundant informationand reducing the amount of transmission, so as to improve the energy and bandwidth efficiency ofthe whole network. Sensors suffer from restricted computation, storage, and power resources, andthey always communicate with each other in an ad-hoc way. Consequently, the process of dataaggregation is under the threat of various kinds of attacks. Attackers may eavesdrop in the air orcompromise legitimate sensor nodes, and inject forged messages into the network in the name ofthat compromised node. Recently, wireless sensor network is popular in many areas, such asmilitary, finance, and medical care. And, the security of data aggregation is of vital importance.A lot of traditional aggregation schemes are designed to support confidentiality or integrity ofthe transmitted data. However, these schemes are of low security level or high energy consumption.To address the problem, we launched our research on the improvement of security level and thereduction of energy consumption. The main contributions of this work are as follows:1. First, we introduce the basic concept of wireless sensor network and data aggregation. Then,some typical aggregation schemes are introduced, and the performance of those schemes areanalyzed and compared with each other.2. Given the knowledge of data aggregation and authentication (DAA) protocol, we proposed aconcealed data aggregation (CDA-FDD) scheme, which is based on false data detection. AlgorithmCDA-FDD is an integration of false data detection and the cluster-based hierarchical sensornetworks, and it can provide end-to-end confidentiality by applying the elliptic curve basedhomomorphic encryption scheme. In the process of data collection, it can detect and discard falsedata which is injected by vicious attackers during the aggregation process. As a consequence, thetransmission overhead is reduced and the accuracy of aggregation is enhanced. The experimentalresult shows that algorithm CDA-FDD improves the security level as well as reducing the energyconsumption of the whole network.3. The recoverable concealed data aggregation (RCDA) scheme can recover row sensed data atthe base station. To address its problem of high transmission overhead, we proposed a novelreference-based recoverable concealed data aggregation (R-RCDA) scheme. Algorithm R-RCDAsets a new reference for every sensor node dynamically, and the reference will be used in the next round of transmission. It can reduce the amount of transmission by transmitting the differencebetween the reference and the row sensed data. The experimental result shows that R-RCDA canreduce the amount of data transmission effectively and then improve the energy and bandwidthefficiency of the whole network.