Pattern Matching in Text Compressed by Using Antidictionaries

In this paper we focus on the problem of compressed pattern matching for the text compression using antidictionaries, which is a new compression scheme proposed recently by Crochemore et al. (1998). We show an algorithm which preprocesses a pattern of length $ m $ and an antidictionary M in $ O(m^2 + left |M\right |) $ time, and then scans a compressed text of length n in $ O(n +r) $ time to find all pattern occurrences, where $ left |M\right | $ is the total length of strings in $ M $ and $ r $ is the number of the pattern occurrences