铁路路基病害检测雷达信号中的强干扰谱分析及滤波处理

Spectrum analysis and filtering on suppressing intensive noise in train-mounted GPR data

铁路路基病害诱发铁路路基失稳,是影响铁路安全运营的重要因素之一。开展快速高效的铁路路基检测方法研究,对铁路日常维护具有非常重要的意义。探地雷达是一种无损高效的检测手段,但用于铁路检测的工作环境是一种强电磁干扰的环境,路基介质的雷达反射信号被轨枕等介质的强干扰反射所湮没。压制此类强干扰是提高铁路路基病害信息识别的主要途径。介绍了多通带滤波原理及实现方法,分析了铁路路基雷达信号中轨枕强反射干扰的频谱特征,并针对其特点设计了多通带滤波器,将车载雷达实测数据进行滤波处理,得到了滤波后的雷达图像。处理结果表明,原始雷达检测信号经滤波处理后,轨枕强干扰信号被压制,提高了路基病害的弱反射信号的信噪比。因此,多通带滤波能有效压制轨枕强干扰信息,便于铁路路基病害地识别。

Railway subgrade defect is one of the most important inducements that leads to potential hazards in railway trans -portation. A rapid and nondestructive detective method is very valuable for routine railway subgrade maintenance. As a nonde-structive detection testing tool, ground penetrating radar （GPR） has attracted great interest in railway subgrade evaluation. However, the radar signal is seriously degrade by the noise and clutters by railway infrastructures. Suppressing the high level interference is essential to identify the subgrade defects. In this paper, we firstly introduced the multi - band filter and proposed the filtering algorithm. Then we analyzed the frequency spectrum of the GPR data containing high level noise and designed the especial filter to process the GPR data. The high level noise is dramatically suppressed and the weak reflection signal from the subgrade defects can be identified.