Abstract: Outlier detection in wireless sensor network(WSN) is of great significance for environmental monitoring.Two outlier detection methods for WSNs were proposed based on BP neural network and linear neural network in this paper.Latest historical data with fixed length of data window was used to train a neural network model,and then these methods can predict the sensor data of the next time.A confidence interval with probability p was calculated with the help of the model residual.The new measurement will be identified as normal one if it falls inside the prediction interval.Otherwise,it will be classified as an abnormal record.In order to compare and demonstrate the performance of the proposed methods,we finished the simulation experiments in Matlab environment.The experiment results show that the detection rate of outlier detection based on linear neural network reaches 97.9%,and the false positive rate is less than 0.76%.While the detection rate of outlier detection based on BP neural network reaches 96.7%,the false positive rate is less than 0.84%.

Key words: BP neural network,Linear neural network,Outlier data detection,Detection rate,False positive rate