摘要
去噪是激光回波处理的重要一环,好的去噪算法可以有效地去除噪声带来的影响,保留有效信号,减少失真。经验模式分解以信号的局部特征时间尺度为依据,将时域信号分解为若干个不同频带尺度的时域信号分量,传统的去噪实现一般是直接舍弃高频分量。然而,由于在某些高频分量中,存在着有效信号和噪声的混叠,如果直接舍弃,会导致有效信号的损失。由此考虑并提出了一种基于经验模式分解及分数阶傅里叶变换的激光雷达回波去噪算法,用分数阶傅里叶变换处理特定的本征模态函数,以此来提高去噪的有效性及准确性。
The laser echo waveform denoising is an important part of the whole process, a good denoising algorithm can remove the bad effects caused by the noise availably, preserve useful information and effectively reduce the information＇ s distortion. Empirical Mode Decomposition （EMD） decomposes the signal into several time domain＇ s components of different band scale. In the traditional EMD denoising, it directly abandons the component of high frequency. In this way, however, the useful information will be lost because in some high frequency components useful information would intermix with noise. In the paper we consider and propose a denoising algorithm based on the EMD and Fractional Fourier Transform（ FRFT）, we choose the specific Intrinsic Mode Function（IMF） and process it with FRFT to improve the final accuracy and precision.
出处
《激光杂志》
北大核心
2016年第11期76-79,共4页
Laser Journal
基金
国家科技重大专项(AHJ2011Z001)
