Attack Detection and Classification of Heterogeneous Wireless Sensors Using Co-Clustering
In a Wireless Sensor Network a large number of sensors are deployed for the purpose of sensing data and then to bring the data back securely to nearby base stations. The base stations then perform the costly computation on behalf of the sensors to analyze the data sensed by the sensors. Due to resource limitations of the nodes and also due to the vulnerability of physical captures of the nodes, the traditional cryptographic techniques are very complex and should not fit for energy constrained environments. Data mining techniques can be applied to find the malicious behavior of the nodes of the Sensor Network.