Under-ice target echo time delay estimation using [formula omitted]-norm based matched filter

Abstract

In the ice-covered underwater environment of the Arctic area, impulsive noise is one of the most challenging problems for target echo time-delay estimation (TDE) and severely degrades the performance of some so-called traditional TDE algorithms. In this work, we propose a TDE approach called “ℓp-norm-based correlation,” which is explored for use in active sonar systems in an impulsive environment. The ℓp-norm-based optimization is presented here to improve the robustness, and we introduce the Toeplitz matrix to formulate the received signal. The different columns of the desired matrix represent the various time-delay replicas of the transmitted signal, and the global minimizer can be found by rewriting the optimization problem. Thus, all global minimizers are efficiently solved simultaneously. The proposed TDE technique can enhance time-delay resolution and provide reliable estimation results compared with conventional matched filtering and other competing approaches. Furthermore, we show that the performance of the resulting method can be verified by the results of numerical simulation and by comparison with the experimental data collected from the 11th Chinese National Arctic Research Expedition.