A New Algorithm for Pattern Match Based on BM Algorithm
Mei Jin-jie
Abstract
Mei Jin-jie
Abstract
To improve the speed and benefit of pattern match in an intrusion detection system,a new improvement of algorithm was proposed,on the basis of analyzing BM algorithm and its related improvement of algorithm,where the merits of BMH and BMHS algorithms were integrated and the uniqueness of the last character of character string and its corresponding to the next character of text string was utilized.Meanwhile,the information of text string was taken into account and the matching times were reduced greatly while the characters of text string were skimmed as many as possible,realizing the fast search of the text string without matching each character of text string.Experimental results show that the improved algorithm can effectively enhance the speed of pattern matching.Compared with BM,BMH and BMHS algorithms,the new algorithm can boost the performance by 9.9%,6.2% and 3.3% in the pattern length is 15,improving the efficiency of the intrusion detection.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
To improve the speed and benefit of pattern match in an intrusion detection system,a new improvement of algorithm was proposed,on the basis of analyzing BM algorithm and its related improvement of algorithm,where the merits of BMH and BMHS algorithms were integrated and the uniqueness of the last character of character string and its corresponding to the next character of text string was utilized.Meanwhile,the information of text string was taken into account and the matching times were reduced greatly while the characters of text string were skimmed as many as possible,realizing the fast search of the text string without matching each character of text string.Experimental results show that the improved algorithm can effectively enhance the speed of pattern matching.Compared with BM,BMH and BMHS algorithms,the new algorithm can boost the performance by 9.9%,6.2% and 3.3% in the pattern length is 15,improving the efficiency of the intrusion detection.
Key concepts: String searching algorithm, Commentz-Walter algorithm, Character (mathematics), Algorithm, Approximate string matching, String (physics), Pattern matching, Intrusion detection system