A Novel Denoise Method of Acoustic Signal from Train Bearings Based on Resampling Technique and Improved Crazy Climber Algorithm

The wayside acoustic defective bearing detector system (TADS) is located on both sides of the railway, so that signals recorded by microphone not only include sound from train bearings but also it other disturbance sources. heavy noise and multisource would badly reduce reliability accuracy detection result TADS. In order to extract useful information signal exactly efficiently, a novel denoising method based Short-time Fourier transform (STFT) improved Crazy Climber algorithm was in this paper. Firstly, STFT performed obtain time-frequency distribution matrix. Based original algorithm, movement rule fitting process ridge lines were presented which could accurately rapidly. way, important be divided signals. Finally, simulation experimental verifications carried out, has proved effective extracting matrix dividing form