一种低压配电网载波通信信号分析的新方法

A NOVEL SCHEME FOR LOW VOLTAGE POWERLINE COMMUNICATION SIGNAL PROCEESSING

低压电力网的噪声成分复杂 ,随负荷变化大 ,且具有很强的非平稳特性 ,因此 ,如何有效地对接收的载波信号进行处理是个十分困难的问题。该文通过分析小波及小波包的特性 ,提出了一种可以解决这一困难的方法。小波分析是一种一维时频分析方法 ,它同时具有时域和频域的良好局部化基金项目 :中华电力教育基金会许继奖教金资助项目。特性及自动调节的时频窗 ,可以聚焦到信号的任意细节 ,但对高频信号的频域分辨率较低。小波包分析能够为信号提供一种更加精细的分析方法 ,它对小波分析中没有细分的高频部分进一步分解 ,并能够根据信号的特征 ,自适应地选择相应的频带 ,使之与信号频谱相匹配 ,从而提高了频率分辨率。该文对二者进行了分析比较 ,并将其应用于处理实际低压配电网中采集到的通信信号 。

The quality of the signal transmission over the low voltage powerline is strongly polluted by the omnipresent noise, heavily jammed due to a variety of loads and unpredictable effects on signal propagation. For this reason, effective signal processing of such kind of signal presents difficulties. This paper aims to find out an effective method to deal with this problem. Wavelet transform, a linear time frequency analysis method, has good characters in both the time and the frequency localization and can give a self adjusting time frequency window. However, When dealing with high frequency signal, it has the drawbacks of low frequency domain resolution. On the other hand, Wavelet packet transform can provide a more accurate analysis. Since it has perfect performances in frequency localization, especially to high frequency signal, one can focalize himself on any details of a given signal and get good result. Application of the proposed scheme to the signal measured from the real low voltage powerline communication system shows the effectiveness of the proposed method.