An improved multiple patterns matching algorithm for intrusion detection

Pattern matching algorithm is one of the Core algorithms in the detection engine of the intrusion prevention system. Efficiency of the intrusion prevention system is determined by pattern matching algorithm. A survey of the pattern matching algorithm is described in this thesis. The Wu-Manber algorithm which is one of the multi-pattern matching algorithm is explained in detail and the improvement of the Wu-Manber algorithm is presented to improve the efficiency. By dividing the pattern group into two subgroups and dealing with the two subgroups in different methods, the QWM algorithm enhanced the efficiency of pattern matching. Experimental results show that when pattern group contains a pattern that is less than three bytes, the performance of the improved Wu-Manber algorithm is much better than the traditional Wu-Manber algorithm.