用于LBG初始码书设计的改进PNN算法  

Improved PNN algorithm for initial codebook design of LBG

作　　者：陈善学[1] 张艳[1] 吴立彬[1] 

机构地区：[1]重庆邮电大学移动通信安全技术实验室,重庆400065

出　　处：《重庆邮电大学学报（自然科学版）》2012年第1期50-54,共5页Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)

基　　金：国家科技重大专项资助项目(2009ZX03001-004);国家自然科学基金(61071116);重庆市科委自然科学基金(CSTC;2010BB2407)~~

摘　　要：矢量量化初始码书对于码书的形成非常重要,为了改善初始码书的性能和提高最终的码书质量,在分析成对最邻近(pair-wise nearest neighbor,PNN)算法与基础上,提出2种改进算法用于LBG算法初始码书的设计。改进的算法将训练矢量的分量和值排序与一次迭代多次融合用于PNN算法中,有效地降低了PNN算法的复杂度,减小了PNN算法的收敛时间。实验证明,该算法具有合理性和有效性,与LBG算法结合可进一步提高码书质量。In vector quantization（VQ）,the initial codebook design is very important for VQ codebook.This paper introduces two improved PNN algorithms for the design of the initial codebook of LBG algorithm in order to promote the performance and quality of final codebook.The improved algorithm uses sorting training vectors and to the sum of each training vector components and merging multiple training vector iteration.The improved PNN algorithm effectively reduces the complexity and convergence time.Experiments show that the algorithm is reasonable and effective.The proposed method can be combined with the LBG algorithm to further improve the quality of codebook.

关 键 词：矢量量化 初始码书 LBG算法 成对最邻近算法(PNN算法) 

分 类 号：TN911.21[电子电信—通信与信息系统]