Genetic algorithm based analysis of envelope using complex analytical wavelet transform

Abstract

Hilbert transform envelope analysis, spectrogram envelope analysis, and complex analytical wavelet envelope analysis are discussed in this paper. According to the disadvantages such as Hilbert transform susceptible to noise interference, spectrum analysis affected by short-time window, and the parameters of Morlet wavelet set manually by experience, an improved algorithm based on genetic algorithm and complex Morlet wavelet transform is put forward in this paper. By genetic algorithm, the parameters of Morlet wavelet are optimized without experience, and the wavelet transform can effectively eliminate the influence caused by random noise. Moreover, the experimental results show that this improved algorithm outperforms the other algorithms.