A Fast Iterative Adaptive Algorithm for Direction of Arrival Estimation of Millimeter-Wave Radar

Iterative Adaptive Algorithm (IAA) is a super-resolution algorithm widely applied for highly accurate estimation of Direction of Arrival (DOA) in millimeter-wave radar. However, traditional IAA encounters challenges related to algorithmic complexity and computational delay, making it unsuitable for applications requiring real-time performance. In order to reduce the angle estimation errors caused by the mismatch between the target location and the grid dictionary, a common approach is to refine the grid. However, it will further exacerbate the slow computation problem of IAA. To tackle this problem, a Fast Iterative Adaptive Algorithm (FIAA) is proposed in this paper. Built upon the traditional IAA, the FIAA adopts a hierarchical grid refinement approach to iteratively calculate target angles. Initially, a coarse grid division is implemented across the entire spatial domain and the potential regions of real target location are identified by utilizing the IAA. Subsequently, within these potential regions, a refined grid division is executed, accompanied by updates to the signal direction matrix. Ultimately, the IAA, incorporating a regularized covariance matrix, is employed to attain highly accurate estimations of the target angles. Experimental results demonstrate that the proposed FIAA can effectively circumvent scanning and computation in non-signal potential regions, reducing computational time to at least 4% of that required by IAA. Moreover, the computational accuracy of FIAA is nearly identical to that of IAA when the SNR exceeds 10 dB, making it suitable for high-real-time, high-accuracy DOA estimation in millimeter-wave radar applications.