等均值等范数最近邻矢量量化码字搜索算法

An Equal-Average Equal-Norm Nearest Neighbor Codeword Search Algorithm for Vector Quantization

本文提出了一种等均值等范数最近邻 (EENNS)矢量量化码字搜索算法 .在编码前 ,该算法预先计算每个码字的均值和范数 ,然后根据均值大小的升序排列对码字进行排序 .在编码过程中 ,首先选取与输入矢量均值最近的码字作为初始匹配码字 ,然后利用两条有效的删除准则在该码字附近进行上下搜索与输入矢量最近的码字 .测试结果表明 ,本文算法比等均值最近邻搜索算法 (ENNS)和最近提出的范数排序搜索 (NOS)

An equal average equal norm nearest neighbor codeword search algorithm(EENNS)is presented for vector quantization in this paper.Before encoding,the mean value and norm are first computed for each codeword,and then codewords are ordered according to the ascending order of their mean values.During the encoding process,the algorithm first selects the codeword that has minimum mean distance from the input vector as the tentative matching codeword,and then applies two efficient elimination criteria to search the nearest codeword close to the tentative matching codeword up and down.Experimental result shows that this algorithm is more efficient than the equal average nearest neighbor search algorithm(ENNS)and recently presented norm ordered search algorithm(NOS).