Feature detection with ultrasonic phase-dependent baseband echo separation

Ultrasonic feature detection is primarily based on the time-of-flight of the acoustic signals. Operating at their resonances, the transducers tend to produce long transients that exacerbate the effects of multi-echo interference and result in a loss of details and accuracy. In this work, a novel phase–dependent baseband model is developed to quantitatively describe the ultrasonic signals and the associated interference characteristics. A Fourier approximation algorithm is then applied to retrieve the timing information by separating the primary and secondary echoes from the received baseband echo signal to enhance the measurement resolution. The performance of the proposed algorithm is demonstrated by both numerical and experimental examples.