Dynamic Trust Evaluation of Wireless Sensor Networks Based on Multi-factor

Abstract

Most of the existing trust evaluation methods of wireless sensor networks focus on communication behaviors in building paths without considering transmission behaviors in sensing environment. Therefore, a potential threat tothe network security exists in wireless sensor networks. Wepropose a dynamic trust evaluation method based on multi-factor including direct trust factor and indirect trust factor in this paper. Firstly, direct trust factor is composed of communication trust factor and transmission trust factor, both of which can be got from observing the interaction process. Secondly, indirect trust factorcomes from other nodes' recommendations, which are classified into certainand uncertain ones. Such two kinds of recommendation trust values are computed respectively. Finally, nodes' trustworthinessis measured by combining direct trust with indirect trust dynamically. Besides, both the involved classification standard and dynamic weight assignment are dependent on the interaction times between nodes, which are put forward under the background of Hoeffding's Inequality in Probability Theory. Experiments on NS-2 platform show that the proposed method is effective in increasing the network throughput and the network packet delivery ratio.