Wavelet Denoising Algorithm for Ultrasonic Ranging

Luo Xiuzhi
2012-01-01
Abstract:Effective de-noising methods are necessary in ultrasonic ranging to improve the measurement precision and stability.A de-noising algorithm based on wavelet transforms is presented in this paper.Wavelet transforms are a time-frequency analysis method that indicates the frequency distribution at each sampling point.This model analyzes the sampled data with a stationary wavelet transform(SWT) and correlates the wavelet coefficients with those of an ideal echo signal.The correlation result was then used to identify and eliminate the noise.This algorithm was evaluated on signals acquired with an in-house ultrasonic ranging system with better performance compared to the classic wavelet threshold method.The signal-noise ratio(SNR) is enhanced by 6 to 9 dB and large amplitude noise is effectively suppressed.
What problem does this paper attempt to address?