A Generalized Noise Reconstruction Approach to Robust DOA Estimation

Abstract

<p>A generalized noise reconstruction approach is proposed for improved direction-of-arrival (DOA) estimation. The objective function kernel of the noise reconstruction problem can be made up of many distinct DOA estimation algorithms, making it an optimization problem. First, a coarse estimation of the received domain noise is calculated based on the cumulative sum of an approximate noise first-order finite difference. In particular, the generalized pattern search algorithm is used to find the best initial value. The objective function is constructed with the total noise power estimated by least squares (LS), minimized. After that, a brand-new cumulative sum constructed with the solution from the previous step is used to precisely reconstruct the received data-domain noise. Finally, DOA estimation is obtained using Capon algorithm. The efficiency of suppressing sidelobes and pseudo-peaks and the accuracy of DOA estimation are evaluated. In simulations, different scenarios, such as non-uniform noise, colored noise, impulsive noise, and K-distribution sea clutter, are considered. Real-world frequency-modulated continuous wave (FMCW) radar multiple-input multiple-output (MIMO) data is used to confirm the effectiveness of the proposed approach.</p>