| 摘要: |
| 稀疏重构算法划分网格数目直接决定计算复杂度大小，且其在低信噪比和小快拍下的离网格参数估计性能仍无法满足实际精度需求。为解决上述问题，提出了一种基于张量的矢量共形阵列（Vector Conformal Arrays,VCA）离网格优化参数估计算法。首先，利用信号空域稀疏特性，基于VCA建立二维稀疏离网格张量接收信号模型；然后，为进一步促进解的稀疏性，提出一种三阶分层先验贝叶斯模型，利用张量变分稀疏贝叶斯学习算法得到波达角度（Direction of Arrival,DOA）估计值。在DOA估计过程中，提出一种离网格优化思想，大大降低运算复杂度提升算法效率。最后，利用最小特征向量方法得到信源极化参数估计。仿真结果表明，与未采用离网格优化的算法相比，所提算法的计算复杂度提升约30.8%；同时，在信噪比小于0 dB和快拍小于150的条件下，所提算法的参数估计精度和角度分辨概率分别提升约35.7%和54.4%。 |
| 关键词: 矢量共形阵列 离网格优化 参数估计 变分稀疏贝叶斯学习 |
| DOI：10.20079/j.issn.1001-893x.240727003 |
|
| 基金项目:国家自然科学基金青年科学基金（61801308）；辽宁省兴辽英才计划项目（XLYC1907195） |
|
| A Parameter Estimation Algorithm for Off-grid Optimization of Vector Conformal Arrays |
| JIANG Lai,WANG Siming,LAN Xiaoyu,WANG Sanxi |
| (1.State-owned Changhong Machine Factory,Guilin 541003,China;2.School of Electronic Information Engineering,Shenyang Aerospace University,Shenyang 110136,China) |
| Abstract: |
| The number of grids divided by the sparse reconstruction algorithm directly determines the size of the computational complexity,and the performance of off-grid parameter estimation under low signal-to-noise ratio(SNR) and small snapshots fails to meet the practical accuracy requirements.To solve above problems,a tensor-based off-grid optimization parameter estimation algorithm is proposed for vector conformal arrays(VCA).Then,to further promote the sparsity of the solution,a third-order hierarchical prior Bayesian model is proposed to obtain the direction of arrival(DOA) estimation using tensor variable sparse Bayesian learning algorithm.In the DOA estimation,an off-grid optimization idea is proposed to greatly reduce the computational complexity and enhance the efficiency of algorithm.Finally,the minimum eigenvector method is used to obtain the source polarization parameter.Simulation results show that the computational complexity of the proposed algorithm is improved by about 30.8% compared with that of the algorithm without off-grid optimization;at the same time,the parameter estimation accuracy and angle resolution probability of the proposed algorithm are improved by about 35.7% and 54.4%,respectively,under the conditions that the SNR is less than 0 dB and the snapshot is less than 150. |
| Key words: vector conformal array off-grid optimization parameter estimation variational sparse Bayesian learning |
《电讯技术》编辑部 