Fault diagnosis for wireless sensor network based on genetic-support vector machine

It is well-known that the correct diagnosis for wireless sensor network can avoid the paralysis of entire systems. Here, fault diagnosis for wireless sensor network based on genetic-support vector machine is presented in the paper. In SVM, inappropriate training parameters can lead to over-fitting or under-fitting. Thus, genetic algorithm is used to select the appropriate training parameters of support vector machine. Genetic algorithm is a kind of evolutionary computing algorithm, which has strong global search ability. In the experiments, 60 state samples of wireless sensor network are employed to study the diagnosis ability of genetic-support vector machine. The experimental results show that the diagnosis accuracy of the genetic-support vector machine model is higher than that of the support vector machine model.