数字用户环路传输码型的分析

Modified Competitive Learning Algorithm for Vector Quantization of Images

对数字用户环路系统的传输码型ＡＭＩ，ＨＤＢ３，ＦＯＭＯＴ，２Ｂ１Ｑ，４Ｂ３Ｔ（ＳＵ３２）等多种码型的功率谱密度、近远端串音信噪比和误码率等方面进行了全面的分析、比较，认为采用２Ｂ１Ｑ码型为数字用户环路系统的传输码型最佳．

Neural network is a very efficient method for Vector Quantization (VQ). We proposed a Modified Competitive Learning Algorithm (MCL) basing on the analyses of Competitive Learning Algorithm (CL) and Frequency Sensitive Competitive Learning Algorithm (FSCL), experimental results show that MCL is better than CL and FSCL.