一种基于快速迭代自适应算法的毫米波雷达波达角高精度测量方法  

A High-Accuracy Direction of Arrival Measurement Method for Millimeter-Wave Radar Based on a Fast Iterative Adaptive Algorithm

作　　者：于丹阳 杜磊[1] YU Danyang;DU Lei(National Institute of Metrology,Beijing 100029,China)

机构地区：[1]中国计量科学研究院,北京100029

出　　处：《计量科学与技术》2024年第11期3-9,21,共8页Metrology Science and Technology

基　　金：国家重点研发计划(2022YFF0604803、2017YFF0205006);国家自然科学基金项目(62131001)。

摘　　要：迭代自适应算法(Iterative Adaptive Algorithm,IAA)是一种超分辨算法,广泛用于毫米波雷达波达角(Direction Of Arrival,DOA)的高精度测量之中。然而,传统的IAA存在算法复杂、计算结果迟滞的问题,难以适用于对实时性要求较高的场景。此外,为了解决信源位置与网格字典不匹配而导致角度测量误差较大的问题,常采用网格细化的方法,这将进一步加剧IAA计算缓慢的问题。针对上述问题,提出了一种快速迭代自适应算法(Fast Iterative Adaptive Algorithm,FIAA)。FIAA采用粗细网格分次测量信源角度。首先在全空域内进行粗网格划分并使用IAA计算出真实信源的潜在区域,然后在信源潜在区域内进行细网格划分并更新信号方向矩阵,最后使用具有正则化协方差矩阵的IAA对信源角度进行高精度测量。实验结果表明,FIAA可以有效避免对非信源潜在区域的扫描与计算,计算耗时至少降低为IAA的4%,并在信噪比高于0dB时与IAA的计算精度基本一致,适用于高实时、高精度的毫米波雷达波达角测量场景之中。The Iterative Adaptive Algorithm(IAA)is a super-resolution algorithm widely applied for high-accuracy direction of arrival(DOA)measurements in millimeter-wave radar systems.However,traditional IAA faces challenges such as algorithmic complexity and computational delays,rendering it unsuitable for real-time applications.Additionally,to mitigate angle estimation errors caused by mismatches between the source locations and the grid dictionary,grid refinement is commonly employed,further exacerbating the slow computational performance of the IAA.To address these issues,this paper proposes a Fast Iterative Adaptive Algorithm(FIAA).The FIAA utilizes a hierarchical grid refinement approach to iteratively estimate source angles.Initially,a coarse grid is applied over the entire spatial domain,identifying potential areas of the actual source locations using IAA.Subsequently,in these identified regions,a refined grid division is applied,and the signal direction matrix is updated.Finally,the IAA,incorporating a regularized covariance matrix,is utilized to achieve high-accuracy angle measurements.Experimental results show that FIAA effectively avoids scanning and computations in non-signal regions,reducing computational time to as little as 4%of the IAA,while maintaining comparable accuracy when the signal-to-noise ratio(SNR)exceeds 0dB.This approach is well-suited for high real-time and high-accuracy millimeterwave radar DOA measurement scenarios.

关 键 词：计量学 毫米波雷达 迭代自适应 波达角测量 网格划分 

分 类 号：TB936[一般工业技术—计量学]