DOA Estimation Based on Coherent Integration-Sparse Bayesian Learning with Time-Variant Gain-Phase Errors
Chaojie Lu,Jibin Zheng,Tianyuan Yang,Rui Chang,Hongwei Liu
DOI: https://doi.org/10.1109/taes.2023.3298757
IF: 3.491
2023-01-01
IEEE Transactions on Aerospace and Electronic Systems
Abstract:Direction of arrival (DOA) estimation is an important topic of array signal processing and channels without gain-phase errors or with time-invariant gain-phase errors are assumed in most of the DOA estimation algorithms. However, with the measured data of a reconnaissance receiver system, we verify that the gain-phase error between channels is time-variation. Based on this fact, a more practical signal model with time-variant gain-phase errors is constructed. To obtain accurate DOA estimation, an algorithm termed as coherent integration-sparse Bayesian learning with time-variant gain-phase errors is (CI-SBLVGPE) devised which has two superiorities, i.e., 1) the CI operation integrates the signal energy into a single snapshot, which decreases the gain-phase errors, makes the compensation possible, and guarantees a low computational complexity; 2) the priori information of the gain-phase errors, source DOA, and noise is used to reduce the computational complexity and improve the robustness. With mathematical analyses, numerical simulations, and real data processing, we validate the constructed model and superiorities of the proposed method.