Malicious Node Detection in Sensor Network Using Autoregression Based on Neural Network
Faculty of Computing, Health and Science
School of Engineering
In this paper we propose a strategy based on antecedent values provided by each sensor for detecting their malicious activity. We compare at each time moment the sensor's output with its estimated value computed by a robust autoregressive neural predictor. In case that the difference between the two values is higher then a chosen threshold, the sensor node becomes suspicious and a decision block is activated.