Cluster Based Anomaly Detection in Wireless LAN

Data mining methods have gained importance in addressing computer network security. Existing Rule based classification models for anomaly detection are ineffective in dealing with dynamic changes in intrusion patterns and characteristic. Unsupervised learning methods have been given a closer look for network anomaly detection. We investigate hierarchical clustering algorithm for anomaly detection in wireless LAN traffic. Since there is no standard datasets available to do research in wireless network, we simulated a wireless LAN using NS-2 and the traces are used to observe the traffic patterns. Our study demonstrates the usefulness and promise of the proposed approach which uses hierarchical cluster based framework for anomaly detection in wireless computer networks to produce low false positive alarm and high detection rate also compared with the real time wireless traffic. This system can help Wireless network management system to quickly identify the attacks, which extends the system administrators security management capabilities and improve the integrity of the information security infrastructures. Keywords— Anomaly detection, Wireless Network, Data mining, Clustering , Wireless LAN Traffic data.