De-noising Analysis for Health Monitoring of Submarine Tunnel by Wavelet Transform

Abstract

The popular existing de-noising methods are compared and analyzed, including wavelet packet decomposition, coefficient shrinkage and modulus maxima method of wavelet transform, which advantages and disadvantages are analyzed and summarized. So their suitable scopes and premise condition are obtained. To accurately analyze high frequency part of signal in submarine tunnel monitoring, the maxima method de-noising method based on wavelet transform is suitable for the single characteristics, which can effectively remove the noise interference, and can be well preserved the original signal high frequency detail of tunnel itself part. Finally, the modulus maxima method are successfully adopted to de-noised the monitoring results, and the data de-noised and reconstructed are reliably provided for healthy diagnosis of submarine tunnel.