Key Node Identification Based on Overall Network Performance Considering Cascading Failure in Wireless Sensor Network

Traditional key node identification methods for wireless sensor network (WSN) use certain characteristics of the network to judge the importance of the node and get the importance order, which will lead to errors with the differences in network structure. Meanwhile, the cascading failure of the network has not been considered. In this paper, we treat the construction of the node sequence as an optimal problem. Firstly a network robustness objective function is constructed by considering the cascading failure characteristic of WSN to measure network performance, the reduction of which will assess the destruction caused by node removal from the sequence and further the importance of these nodes. Then a construction model of node sequence is established with the goal to minimize objective function, and finally artificial fish swarm algorithm(AFSA) is used to solve the model. The simulation study indicates that the proposed method performs better than the previous ones and it is applicable to different network structures. Furthermore, the AFSA is more effective than the traditional intelligent algorithms already used in the key node identification.