A Robust and Highly Accurate Super-Resolution Time Delay Estimation Method in the Underwater Acoustic Channel

In the underwater acoustic (UWA) channel, the time-delay estimation is a critical issue in the underwater acoustic measurements, localization, detection, and so on. However, the available bandwidth of UWA signals is narrow, the delay estimation accuracy of conventional matched filtering method is limited. Many existing time-delay super-resolution estimation algorithms perform poorly due to high computational complexity and poor robustness. In this paper, a modified Hybrid-WRELAX method is proposed to deal with the above problem. By changing the initialization parameters and the convergence conditions of the cost function, the computational complexity of the algorithm is reduced, and using the multi-parameter data fusion method to improve the robustness in the actual experimental data processing. Numerical simulations show that the modified method can resolve multipath time delay less than 0.3 times the Rayleigh limit, and the delay estimation accuracy is close to the maximum likelihood estimation in the AWGN environment. Finally, the experimental data analysis is presented to demonstrate the usability of the proposed algorithm.