Application of fuzzy window Markov chain in IDS
模糊窗口Markov链在IDS中的应用

The traditional static Markov model is based on such premise of assumptions as transition probability distribution of system mode of t+1 moment is only interrelated with the state at time t but not with that before time t, which brings big error. Therefore, a new window Markov chain was put forward, and fuzzy measure was introduced into it. The experiment confirms that this model has a good discrimination to the normal behavior and the unusual behavior, and has a faster calculation speed, and it is suitable for the on-line detection.