算术编码的附加位技术及其应用

算术编码是一种高效的熵编码，但是算术编码的非分组性阻碍了算术编码的灵活应用，深入分析算术编码的计算机实现算法，发现用有限长寄存器表示无限长小数，其编码输出流是可以分割的，提出了实用的算术编码附加位技术，可以在编码输出流的恰当位置嵌入附加信息而不影响解码的正确，给出了这种技术的两个应用实例，易于用单片机实现的心电数据无损压缩方法以及对JPEG图象压缩方法的改进。

The Appendant Bits Technique of Arithmetic Coding and Its Application

Arithmetic Coding is a highly effective entropy coding. But, its indivisibility obstructs its flexible application. The computer method of realiging Arithmetic Coding is deeply analyzed and it is found that the bits flow of Arithmetic Coding can be divided, provided the length of the register used in coding process is limited. A practical appendant bits technique of Arithmetic Coding is proposed. The appended bits can be embeded at the suitable position of coding data flow while its decoding is still correct. Two examples are shown: the ECG data compression using single chip microcomputer and the improvement of JPEG based image compression.