| AWGN信道中盲源分离的时频联合两步消噪预处理方案 |
| |
| 引用本文： | 王尔馥,张乃通,孟维晓. AWGN信道中盲源分离的时频联合两步消噪预处理方案[J]. 哈尔滨工程大学学报, 2009, 30(4) |
| |
| 作者姓名： | 王尔馥 张乃通 孟维晓 |
| |
| 作者单位： | 哈尔滨工业大学,通信所,黑龙江,哈尔滨,150080;哈尔滨工业大学,通信所,黑龙江,哈尔滨,150080;哈尔滨工业大学,通信所,黑龙江,哈尔滨,150080 |
| |
| 基金项目： | 国家重点基础研究发展规划(973计划) |
| |
| 摘 要： | 针对加性高斯白噪声AWGN环境下盲源分离算法的性能随信噪比降低而大幅下降的难点问题,借助于经验模态分解和小波变换2种时频分析方法,并对其消噪性能进行优势互补,采用"粗消噪-细消噪"的两步化思想,提出一种应用于盲源分离的时频联合两步消噪预处理方案.仿真试验与其他算法进行对比,证明了新的噪声预处理方案可以更有效地降低观测信号中的加性噪声,使盲源分离算法工作在更宽的信噪比范围内提高算法的鲁棒性,并说明了算法对通信信号具有数据延拓的必要性.
|
| 关 键 词： | 盲源分离 消噪预处理 小波变换 经验模态分解 |
A concurrent time-frequency and noise-pretreatment methodology for blind source separation in an additive white Gaussian noise channel |
| |
| WANG Er-fu,ZHANG Nai-tong,MENG Wei-xiao. A concurrent time-frequency and noise-pretreatment methodology for blind source separation in an additive white Gaussian noise channel[J]. Journal of Harbin Engineering University, 2009, 30(4) |
| |
| Authors: | WANG Er-fu ZHANG Nai-tong MENG Wei-xiao |
| |
| Affiliation: | Communication Research Center;Harbin Institute of Technology;Harbin 150080;China |
| |
| Abstract: | The blind source separation(BSS) problem in additive white Gaussian noise(AWGN) is known as a hard problem;performance of the separation algorithm degrades significantly with decreases in the SNR.Empirical mode decomposition(EMD) and wavelet transform help solve the problem.To combine the advantages of these two methods,the idea of "rough cancellation and then exact cancellation" was adopted and a concurrent time-frequency and noise-pretreatment methodology for BSS was proposed.Simulation results showed tha... |
| |
| Keywords: | blind source separation noise-pretreatment wavelet transform empirical mode decomposition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
