论文部分内容阅读
分析了传统LZW编码算法的不足,根据语音数据的特点,对传统LZW编码数据压缩算法进行了改进,将字典初始化为16位,采用散列法和拉链法进行词条检索,采用阈值判断和LRU淘汰机制改进条目更新的方式,编码时采用自适应变码长方式。经测试,相比于传统LZW编码数据压缩算法,改进的算法对不同码长的数据的适应性更好,并且压缩比提高了约8%。
The disadvantages of the traditional LZW coding algorithm are analyzed. Based on the characteristics of speech data, the traditional LZW coding data compression algorithm is improved, and the dictionary is initialized to 16 bits. Hedge method and zipper method are used to search the entries. Threshold judging and LRU Elimination mechanism to improve the way the entry is updated, adaptive encoding variable length encoding. Tested, compared to the traditional LZW coded data compression algorithm, the improved algorithm for different data length code more adaptive, and the compression ratio increased by about 8%.