A novel method for Estimating Time of Flight of ultrasonic echoes through short-time Fourier transforms

The accurate estimation of time of flight (TOF) of an ultrasonic signal is essential in ultrasonic non-destructive testing (NDT). A novel method for TOF estimation based on the short-time Fourier transform (STFT) is proposed in this paper. Firstly, the time and frequency components of ultrasonic echoes are represented by a STFT. Then, the STFT spectrum of the echoes is projected onto the time domain. Finally, the TOF is estimated by calculating the location of the maximum modulus of STFT spectrum of the signal. A nonlinear Least Squares (LS) estimation method based on Modified Gauss–Newton (MGN) is employed to improve the estimation accuracy. The TOFs of multi-component echo signals are estimated by successive windowing. Numerical simulations and experiment tests show the performance of the proposed method.