High precision time-delay estimation algorithm based on high-order cumulant

Aiming at the problem of low accuracy of time delay estimation in ocean exploration due to low signal-to-noise ratio (SNR) of received signals and various noise disturbances, a new high-order cumulant-based high-precision time delay estimation method called SC-HOCS was proposed here. Firstly, autocorrelation and cross-correlation were conducted for received signals of two channels to suppress part of Gaussian noise. Then signals were processed with the high-order cumulant one-dimensional slice method to suppress correlated Gaussian noise and non-Gaussian color one, and improve the SNR of received signals. Finally, Hilbert transform was used to sharpen correlation peaks, and further improve the accuracy of time delay estimation. It was shown that compared with the generalized correlation method, quadratic correlation one and high-order cumulant one-dimensional slice one, the proposed method can suppress correlated noise very well and obtain better time delay estimation accuracy under lower SNR; this algorithm has less amount of computation to meet the requirements of real-time data processing; the effectiveness of the proposed method is verified with computer simulation and pool tests; this method provides a new technical way for high precision time delay estimation of signals with low SNR in ocean exploration. © 2020, Editorial Office of Journal of Vibration and Shock. All right reserved.