Direction Of Arrival Estimation in the Presence of Imperfect Waveforms for Multiple Targets in MIMO radar

Abstract

Direction of Arrival (DOA) estimation of multiple targets remains a challenging task with wide-ranging applications in Multiple Input Multiple Output (MIMO) radar systems. While subspace-based methods enhance accuracy, they demand multiple snapshots and elevate computational complexity. This paper presents a novel approach using Wild Horse Optimization (WHO) for DOA estimation in co-located MIMO radar encountering imperfect waveforms, requiring only a single snapshot. The method integrates the theory of extended array manifold vectors into a mean squared error (MSE) framework to craft an objective function that minimizes MSE deviation, guided by a penalty function reflecting discrepancies between actual and desired system responses. Rigorous validation via Monte Carlo simulations—employing histograms, box plots, cumulative distribution function (CDF) graphs, RMSE, noise robustness, and accuracy metrics—demonstrates the effectiveness of the WHO-based approach. Comparative analysis with state-of-the-art algorithms further substantiates its superiority.