On the On-line Arithmetic Coding Based on Antidictionaries with Linear Complexity

This paper proposes an on-line data compression based on antidictionaries with linear time. The proposed algorithm works using only suffix trees without constructing of antidictionaries, and we prove that the time complexity of this algorithm is linear with respect to the string length. Furthermore, the proposed algorithm produces the tree model based on antidictionaries. This tree model gives an efficient probabilistic model for entropy codings. Its effectiveness is demonstrated by simulation results.