FM-to-AM Effect Removal Technology of Pulse Waveform Data Based on Deep Learning

Abstract

In the process of laser pulse transmission, amplification and sampling measurement, the amplitude-to-frequency modulation (FM-to-AM) effect caused by spectral distortion leads to peak oscillation of waveform measurement data, which is difficult to be completely eliminated by improving measurement methods. In this paper, a method of removing the FM-to AM effect of pulse waveform data based on deep learning is proposed. After the original waveform data is de-modulated, the accuracy of waveform prediction is obviously improved. This technology can precisely remove modulated signals while retaining the key features of original data, and can well deal with various complex waveform data.