Abstract: A fast encoding algorithm based on the Mean Square Error（MSE） distortion for vector quantization is introduced.Using the wavelet transforms peculiarity，the vectors can be effectively constructed.It simplifies the use of the Non-Linear Interpolated Vector Quantization（NLIVQ） technique and makes the Partial Distance Search（PDS） algorithm more efficient.Using the relation of vector’s L2-Norm and its Euclidean distance，we obtain some conditions of eliminating unnecessary codewords.During the search process for code，a lot of codewords can be rejected by the proposed algorithm which combines with the Non-Linear Interpolated Vector Quantization technique and the Partial Distance Search technique.Experimental results show that the reduction of computation is outstanding in the encoding time and complexity against full search method.

Key words: fast encoding, Mean Square Error（MSE）, wavelet transform, Vector Quantization（VQ）

摘要： 研究了一种基于均方误差（MSE）测度的矢量量化快速编码算法。算法利用小波变换的特点，合理地构造矢量，便于非线性插补矢量量化技术的使用，也使部分失真排除法的效率大大提高。使用矢量的二范数和距离测度关系的码字排除方法，再结合非线性插补矢量量化技术和部分失真排除法，在搜索编码过程中，有效排除部分候选码字。实验结果表明，相对于穷尽搜索方法，计算量有明显降低，计算时间显著减少。

关键词: 快速编码, 均方误差, 小波变换, 矢量量化