MPEG-4 Audio Lossless Coding(ALS)編碼效應之探討

本論文主要為研究MPEG-4 ALS的編碼特性，因為其最佳編碼是不公開的，且它可快速地為每個區塊決定最佳的預測階數，故我們試著猜測最佳編碼的方法。首先，我們實驗了編碼器的特性與暫態訊號對於區塊變換造成的影響，並實驗了編碼熱門音樂與古典音樂時，在區塊選擇上的差異；而量化的部份，我們為了改善量化造成的誤差實作了方法，另一方面則依最佳化理論去尋找最佳的量化位階，來改善編碼效果；最後，我們從數位信號處理的相關書籍中，得知最小平方法(Least Square method)是另一個可以用來求預測係數的方法，故我們實作了最小平方法與Levinson-Durbin比較，並討論此方法對於本系統是否合適。

關鍵字

MPEG-4音訊無失真編碼；最小平方法；最佳化理論

並列摘要

In this thesis, we study the features of MPEG-4 ALS encoder, the optimal encoder is not public, and it can quickly estimate an optimal order for each blocks, so we try to guess the method of the optimal encoder. First, we test the features of encoder and transients for the choice of block switch level, then experiment the different between pop music and classical music. Furthermore, in order to reduce quantization error, we implement a method, on the other hand, according to the optimization theory, we try to find the best quantization step, to improve the encoding effect. Finally, we found “Least Square method” on digital signal processing books, it can also result the linear prediction coefficients, so we implement it to compare with Levinson-Durbin for residual energy and residual encoded bits, and discuss whether it suits this system.

並列關鍵字

MPEG-4 ALS ； Least Square ； Optimization

參考文獻

[9] 鄭凡寓，音訊編碼空間定位評估系統，碩士論文，國立台北科技大學資訊工程研究所，台北，2008。

[4] T. Liebchen and Y. A. Reznik, “MPEG-4 ALS: an Emerging Standard for Lossless Audio Coding,” IEEE Data Compression Conference, 2004, pp. 439-448.

[5] I. Tabus, Y. Yang and F. Ghido, “Interleaved Optimization-Quantization of Reflection Coefficients: OQ-Levinson-Durbin and OQ-Burg Algorithms,” Institute of Signal Processing, Tampere University of Technology, Finland

[6] B. Porat, A Course in Digital Signal Processing, New York: John Wiley & Sons, Inc., c1997

[7] S. Haykin, Adaptive Filter Theory 2nd ed., Englewood Cliffs, N.J.: Prentice Hall, c1991, pp. 203-205.

國際替代計量

MPEG-4 Audio Lossless Coding(ALS)編碼效應之探討

全文下載

主題瀏覽