基於混沌與正規化條件 Bi-gram 機率之同時壓縮與加密方法

Abstract

本文提出一種全新的基於混沌之同時壓縮與加密的方法。此新方法成功的改善了現有方式之執行時間跟壓縮效能。現有的同時壓縮與加密方法主要有兩種，一種是透過更新查詢表格中的元素來達到提高壓縮率的目的，另一種則是透過利用不同方式來代表密文來完成前述目標，本文提出的新方法跳脫這兩種思維，採用多表格的方式來達到提升效能的目的。為提升系統功能，我們首先提出連續兩個元素的正規化條件 Bi-gram 機率(NCBP)來建構前述之多表格內容。因為這種較高維度的機率模型能夠更準確地描述輸入檔案的特性，所以運用此種高維度的機率模型可以同時減少處理時間以及密文長度。最後，實驗證實本文所提出的新方法的確較現有方法擁有更高的壓縮效能及更快的執行時間。

A new approach to enhance the compression performance and execution time of chaos-based joint compression and encryption schemes is proposed. Instead of finding a new method to update the Look-up Table (LUT) in each iteration or using a different method to represent the ciphertext, we use multiple LUTs based on the conditional bi-gram probabilities of two consecutive source symbols occurring in the input. This high order probability model can represent the characteristics of whole input file more accurately. As a result, it will further decrease the search time as well as the length of the resultant ciphertext. Simulation results also verify that the proposed approach does lead to better compression performance and execution efficiency.