基于多分辨率分析的小波系数压扩去噪算法  

Wavelet Coefficients Companding De-noising Algorithm Based on Multiresolution Analysis

作　　者：王振朝[1] 岳莹昭[1] 师洁[2] 侯惠然[1] 

机构地区：[1]河北大学电子信息工程学院,河北省保定市071002 [2]中国人民解放军92914部队,海南省临高市571800

出　　处：《中国电机工程学报》2008年第10期76-81,共6页Proceedings of the CSEE

基　　金：河北省科技攻关计划项目(06213552)。

摘　　要：研究利用小波分析技术抑制突发性噪声的算法,以改善多载波调制的电力线通信系统中的同步性能。首先,对小波强制去噪算法及其局限性进行分析,提出了基于多分辨率分析的小波系数压扩去噪新算法。新算法在时域上对含噪信号进行分解,在频域上对各层的小波分解系数的畸变量进行抑制。然后,用低压电力线上的实测噪声对新算法进行了仿真验证。结果表明,新算法的去噪效果大大优于现有算法,尤其适用于平稳信号中突发性噪声的抑制。最后,将算法应用于作者开发的基于相对值判决的多载波调制系统中,试验结果证明,新算法可有效提高多载波调制系统的同步性能。This paper studies the algorithm restraining out-bursting noise to improve synchronous performance of multi-carrier modulation （MCM） in power-line communication system. Through analyzing the wavelet forced de-noising （WFDN） algorithm and its limitation, a new algorithm named wavelet coefficients companding de-noising （WCCDN） algorithm is proposed based on multi-resolution analysis. The new algorithm decomposes noisy signal in time phase and suppress deform quantum of wavelet decomposition coefficients in frequency phase. Then the performance of the new algorithm is simulated with noises of real low-voltage power line. The simulating results show that the new algorithm is superior to the existing algorithm especially in mitigation of the out-bursting noise on stationary signal. Finally, the algorithm is applied in the MCM system based on PSD relative judgment and the synchronous performance of MCM system is improved.

关 键 词：小波变换 多载波调制 去噪 多分辨率分析 压扩算法 突发性噪声 

分 类 号：TN914[电子电信—通信与信息系统] TN911[电子电信—信息与通信工程]