Embedding adaptive arithmetic coder in chaos-based cryptography  

作　　者：李恒建 张家树 

机构地区：[1]Sichuan Province Key Lab of Signal & Information Processing,Southwest Jiaotong University

出　　处：《Chinese Physics B》2010年第5期121-129,共9页中国物理B（英文版）

基　　金：Project supported by the National Natural Science Foundation of China (Grant No. 60971104);the Basic Research Foundation of Sichuan Province,China (Grant No. 2006J013-011);the Outstanding Young Researchers Foundation of Sichuan Province,China (Grant No. 09ZQ026-091)

摘　　要：In this study an adaptive arithmetic coder is embedded in the Baptista-type chaotic cryptosystem for implementing secure data compression. To build the multiple lookup tables of secure data compression, the phase space of chaos map with a uniform distribution in the search mode is divided non-uniformly according to the dynamic probability estimation of plaintext symbols. As a result, more probable symbols are selected according to the local statistical characters of plaintext and the required number of iterations is small since the more probable symbols have a higher chance to be visited by the chaotic search trajectory. By exploiting non-uniformity in the probabilities under which a number of iteration to be coded takes on its possible values, the compression capability is achieved by adaptive arithmetic code. Therefore, the system offers both compression and security. Compared with original arithmetic coding, simulation results on Calgary Corpus files show that the proposed scheme suffers from a reduction in compression performance less than 12% and is not susceptible to previously carried out attacks on arithmetic coding algorithms.In this study an adaptive arithmetic coder is embedded in the Baptista-type chaotic cryptosystem for implementing secure data compression. To build the multiple lookup tables of secure data compression, the phase space of chaos map with a uniform distribution in the search mode is divided non-uniformly according to the dynamic probability estimation of plaintext symbols. As a result, more probable symbols are selected according to the local statistical characters of plaintext and the required number of iterations is small since the more probable symbols have a higher chance to be visited by the chaotic search trajectory. By exploiting non-uniformity in the probabilities under which a number of iteration to be coded takes on its possible values, the compression capability is achieved by adaptive arithmetic code. Therefore, the system offers both compression and security. Compared with original arithmetic coding, simulation results on Calgary Corpus files show that the proposed scheme suffers from a reduction in compression performance less than 12% and is not susceptible to previously carried out attacks on arithmetic coding algorithms.

关 键 词：CHAOS CRYPTOGRAPHY compression arithmetic coding 

分 类 号：O415.5[理学—理论物理] TP309.7[理学—物理]