An Extension of DTFT-Based Sinusoidal Signal Time Delay Estimation Algorithm for Linear Time-Varying Situation

To estimate the propagation delay between two noisy sinusoidal signals with known same frequency but different amplitudes, a time delay estimation algorithm based on DTFT is extended in the paper. By using sliding window recursive discrete time Fourier transform (SWRDTFT), the DTFT-based algorithm is able to be used for tracking time delay of sinusoidal signal. This algorithm does not require initial value, which is rather different from traditional adaptive tracking algorithms. By choosing the window size N, the time delay in center of the window is determined. Theoretical derivation shows that this algorithm can achieve approximately Cramer-Rao lower bound (CRLB) for complex sinusoidal signal and asymptotically achieve CRLB for real sinusoidal signal in the sense of N size window under slowly linear time-varying scenario. Simulations results confirm the efficiency and superior performance of the DTFT-based algorithm as compared to the existing filter-based approaches.