| 摘要: |
| 互质空间步进频雷达通过构建虚拟孔径并合成等效大带宽，有效提升了角度与距离分辨率。然而，传统方法通过对样本协方差矩阵进行向量化来获取虚拟阵列数据时，会破坏接收信号在发射-接收-快拍维的固有结构，导致信息损失。尽管基于张量平行因子分解的方法能够保留信号的三维结构，但其分解误差会在后续的互质阵虚拟化过程中被放大，从而影响参数估计的性能。为此，本文提出了一种基于Hankel矩阵重构的稳健参数估计方法。该方法首先构建接收信号的三阶张量模型，并引入了快拍维低秩近似以降低计算复杂度；随后，针对张量分解误差的放大问题，将虚拟导向矢量的估计值重构为Hankel矩阵，并结合截断奇异值分解，有效修正了导向矢量的估计偏差。仿真结果表明，在低信噪比和小快拍条件下，所提方法相较于2D-SSMUSIC、PARAFAC以及HOSVD-RDMUSIC算法，具有更高的估计精度和鲁棒性，且计算复杂度更低。 |
| 关键词: 互质空间步进频雷达 平行因子分解 Hankel矩阵重构 截断奇异值分解 误差抑制 |
| DOI： |
| 分类号:TN957;TN958 |
| 基金项目:国家自然科学基金(No. 62271386) |
|
| A Robust Parameter Estimation Method for Coprime Spatial Stepped-Frequency Radar |
|
|
|
|
| Abstract: |
| The coprime spatial stepped-frequency radar effectively enhances both angular and range resolution by constructing a virtual aperture and synthesizing an equivalent large bandwidth. However, conventional methods that vectorize the sample covariance matrix to obtain virtual array data distort the inherent structure of the received signal across the transmit-receive-snapshot dimensions, leading to information loss. Although methods based on parallel factor decomposition (PARAFAC) preserve the three-dimensional structure of the signal, decomposition errors are amplified during subsequent coprime array virtualization, which negatively impacting parameter estimation performance. To address this issue, this paper proposes a robust parameter estimation method based on Hankel matrix reconstruction. The method begins by constructing a third-order tensor model of the received signal, with a low-rank approximation applied along the snapshot dimension to reduce computational complexity. Then, to mitigate the amplification of tensor decomposition errors, the estimated virtual steering vectors are reconstructed into Hankel matrices and refined via truncated singular value decomposition(TSVD), effectively correcting estimation bias. Simulation results demonstrate that, under low signal-to-noise ratio and small snapshot conditions, the proposed method outperforms 2D-SSMUSIC, PARAFAC, and HOSVD-RDMUSIC algorithms in terms of estimation accuracy and robustness, while significantly reducing computational complexity. |
| Key words: coprime spatial stepped-frequency radar parallel factor decomposition (PARAFAC) Hankel matrix reconstruction truncated singular value decomposition(TSVD) error mitigation |