| 摘要: |
| 针对收发共址多输入多输出（Multiple-Input Multiple-Output,MIMO）雷达的低计算复杂度波达方向（Direction of Arrival，DOA）估计问题，提出一种降维的MIMO雷达高精度DOA新算法。首先采用经过白化的降维矩阵对MIMO雷达脉冲压缩后的接收信号进行降维；然后通过最优信号子空间拟合对无幅度误差阵列流型下的接收信号矩阵进行重构；接下来通过酉变换得到实值增广数据矩阵，并在实值稀疏字典矩阵下对其进行稀疏表示；接着将DOA估计问题转化为行稀疏矩阵的稀疏恢复问题，通过改进的稀疏贝叶斯学习对其进行求解，实现目标DOA的估计。理论分析和仿真实验结果验证了该方法的有效性和实用性。 |
| 关键词: 多输入多输出雷达 波达方向估计 降维 稀疏贝叶斯学习 酉变换 |
| DOI： |
|
| 基金项目:西藏自治区重点学科理论物理建设项目 |
|
| Unitary sparse Bayesian learning for MIMO radar angle estimation using reduced-dimension transformation |
| LEI Wenhua,LI Yong,HU Haibing |
| (College of Science,Tibet University,Lhasa 850011,China) |
| Abstract: |
| Direction of arrival(DOA) estimation with low computational complexity for collocated multiple-input multiple-output(MIMO) radar is considered.A novel high precision DOA estimation method using reduced-dimension transformation is proposed.First,the MIMO radar output signal after pulse compression is transformed into a lower dimensional signal using a whitened reduced-dimension matrix.Second,the receiving signal matrix spanned by the error-free array manifold is reconstructed via optimal signal subspace fitting.Third,the real-valued extended data matrix is formed using unitary transformation and then expressed with a real-valued sparse dictionary matrix.Then,the DOA estimation problem is transformed into a problem of row-sparse matrix recovery.Next,the problem is solved by the proposed improved sparse Bayesian learning method to get targets′ DOAs estimations.The effectiveness and practicality of the proposed method is confirmed by theoretical analysis and simulation results. |
| Key words: multiple-input multiple-output(MIMO) radar direction of arrival reduced-dimension sparse Bayesian learning unitary transform |
