- 學位論文
變動長度碼其利用自我同步字串之循序最大事後機率解碼之研究
A Study of the Sequential MAP Decoding for Variable Length Codes with Self synchronization Strings
若您是本文的作者，可授權文章由華藝線上圖書館中協助推廣。
摘要
在過去的研究中指出，霍夫曼編碼其壓縮效果良好，因此廣泛地被影像或視訊的壓縮標準所採用。使用此類變動長度碼對獨立訊源符號做編碼，其所對應的符號格狀圖架構於2000年首次被提出。在解碼時，利用此格狀圖及斐特比演算法(viterbi)可估算出符號序列的最大事後(maximum a posteriori; MAP)機率，然而霍夫曼碼存在著錯誤蔓延(error propagation)的隱憂。 經由霍夫曼編碼過後的位元串流(bit stream)中,有某些字串本身具有自我同步能力，稱之為自我同步字串(self-synchronization string; SSS)，文獻[1]提出利用自我同步字串的特性與最大事後機率解碼結合，可減輕錯誤蔓延達到提高解碼的效能與減少計算的複雜度。
並列摘要
In the past study , Huffman coding has good capability of compression and it have been using to image or video compression widely. The symbol-level trellis has been proposed to correspond to variable length code which coded independent source symbol in 2000 . In the decoding ,it can use trellis and viterbi algorithm to estimate the maximum a posteriori probability of symbol sequence, but there is error propagation in the Huffman code.The string having self-synchronization capability in the bit stream after Huffman coding is called self-synchronization string. In[1],this paper proposed the method of maximum a posterior probability decoding with self-synchronization string to achieve improvement in both decoding performance and complexity reduction.
參考文獻
[1] L. Cao, L. Yao, and C. W. Chen, “MAP Decoding of Variable Length Codes with Self-Synchronization Strings,” IEEE Transactions on Signal Processing, vol. 55, no. 8, pp. 4325-4330, Aug. 2007.
[2] L. Cao, L. Yao, and C. W. Chen, “Joint Source and Channel Decoding with Symbol Re-alignment,” IEEE Communications Letters, vol. 10, no. 10, pp. 719-721, Oct. 2006.
[3] N. Demir and K. Sayood, “Joint source/channel coding for variable length codes,” in Proc. IEEE Data Comp. Conf., 1998, pp. 139–148.
[4] Y. Yang, W. Yi, Y. Chen, and J. Liu, “Iterative joint source channel decoding in wireless sensor networks,” International Conference on Communications, Circuits and Systems(ICCCAS 08'), May 2008, pp,109-113.
[5] Y. Yang, W. Yi, and Y. Chen, “Reliable transfer of variable-length coded correlated data: An low complexity iterative joint source-channel decoding approach,” in Proc. of IEEE 6th Consumer Communications and Networking Conference (CCNC 09'), Jan. 2009, pp. 1-5.