期刊名称:International Journal of Hybrid Information Technology
印刷版ISSN:1738-9968
出版年度:2013
卷号:6
期号:4
出版社:SERSC
摘要:Most of the current string matching algorithms behave slowly when the amount of patterns increases. In this paper a fast matching algorithm named SSEMatch was designed. PHADDW instruction from SSE (Streamed SIMD Extension) set was used in SSEMatch to produce data confusion, by which the patterns can be distributed into pseudo hash address such that there will be less patterns left for verification matching. With the help of PHADDW, the whole matching time was reduced. Our SSEMatch holds a O(n/m) complexity. Experiment shows that similarly to WM algorithm, SSEMatch performs better when the length of the shortest pattern increases. Also when the amount of patterns increases SSEMatch performs better than WM