Time-of-flight Extraction Method Based on Cross-correlation for Stress Measurement

Abstract

The basic principle of ultrasonic-based absolute stress detection is to estimate the stress by detecting the propagation time of ultrasound in the medium. Because the change of wave speed caused by stress is very weak, how to obtain the propagation time of ultrasonic waves in the medium stably is the key problem. The commonly used time-of-flight (TOF) extraction methods include the peak value method or the threshold method, these methods are very sensitive to noise and waveform distortion in the signal, and have poor stability when the echo's signal-to-noise ratio (SNR) is low. Therefore, we adopt a TOF extraction method based on wavelet noise reduction algorithm and cross-correlation method. The SNR of the echo is improved by wavelet noise reduction, and the TOF is obtained by the cross-correlation method. Moreover, the acquisition mode of one transmitter and two receivers is adopted to solve the reference signal consistency problem in the cross-correlation method. The experimental results show that compared with the peak value method, the cross-correlation method improves the accuracy and stability of TOF extraction.