An Anomaly Detection System Based on Hide Markov Model for MANET
Ye Xia, Junshan Li, Yanling Li · 2010
MANET is highly vulnerable to be attacked because of its special structure. Security measures, such as authentication and encryption, can be used for reducing the possibilities of attacks. But they are far from sufficiency. Intrusion detection is the second mechanism to detect the attacks. Anomaly IDS can be used to detect the abnormal behaviors of the network to find some unknown attacking patterns. This paper proposed an anomaly detection method using a simple Hide Markov Model to detect the abnormal network behaviors. We build a Hide Markov Model by supposing that all the network behaviors are completely normal. If observation sequence does not match the model, we conclude that the behavior is not normal, that is to say, there is an attack.