Intrusion detection method for wireless sensor networks（WSNs） is current research hot spot.On the basis of introductions of the principle of support vector machine（SVM）

WSNs intrusion types and characteristics of different types of data

an intrusion detection system for WSNs based on SVM is put forward.This system divides network topology into three layers structure

which are cluster members

cluster header and Sink node

and each layer can execute judgment of intrusion detection according to SVM training results.The experimental results show that the detection system has characteristics of higher detection rate and low energy consumption.