A computationally efficient algorithm for ultrasonic signal decomposition and flaw detection

Chirp pulses are often used as an effective technique for ultrasonic imaging in NDE or flaw detection applications. Chirp echoes characterize dispersive media while enabling pulse compression and improved echo detection. Characterizing defects in large grain materials is a complex and challenging problem in ultrasonic imaging applications. The complexity is caused by microstructure scattering which often results in interference and masking of ultrasonic flaw echoes. Therefore methods to unravel these complex signals are advantageous. In this investigation, a computationally efficient method for parameter estimation of ultrasonic signals, consisting multiple chirplet echoes is introduced. Analytical results show that the algorithm is efficient and successful in signal representation.