|
|
| |
|
|
| 本文已被：浏览 19次 下载 11次 |
码上扫一扫！ |
|
|
| 基于稀疏贝叶斯学习的RIS系统相位误差自校准DOA估计方法 |
|
吴赟俊, 魏爽
|
|
上海师范大学 信息与机电工程学院, 上海 201418
|
|
| 摘要: |
| 在可重构智能表面（RIS）辅助到达方向（DOA）估计中，RIS上的相位误差与信号相干性共同作用，降低了导向矩阵偏移和协方差矩阵的秩，从而显著影响DOA估计的精度.为此，本文提出了一种基于稀疏贝叶斯学习（SBL）的DOA估计方法.在稀疏贝叶斯学习框架中引入相位误差的联合建模与精度估计，利用Gamma分布表征RIS上相位误差的不确定性，并结合期望最大化算法迭代优化后验概率，实现了对RIS相位误差的有效校正与DOA的精准估计.此外，为解决信号相干性问题，引入酉变换操作以恢复协方差矩阵的秩，从而提升算法在信号相干时的去相关能力，有效提高了算法在密集信号环境中的分辨精度.仿真实验结果表明，所提方法在存在相位误差和信号高度相干的情况下，显著提升了DOA估计的精度和角度分辨. |
| 关键词: 到达方向（DOA）估计 可重构智能表面（RIS） 相位误差 稀疏贝叶斯学习（SBL） 相干信号 |
| DOI：10.20192/j.cnki.JSHNU（NS）.2025.02.009 |
| 分类号:TN911.7 |
| 基金项目:上海市自然科学基金（19ZR1437600） |
|
| RIS-based phase error self-calibration DOA estimation method using sparse Bayesian learning |
|
WU Yunjun, WEI Shuang
|
|
College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, China
|
| Abstract: |
| In reconfigurable intelligent surface （RIS） assisted direction of arrival （DOA） estimation， the combined effects of phase errors on the RIS and signal coherence lead to steering matrix deviation and reduction in the rank of covariance matrix， which significantly impacted the accuracy of DOA estimation. To address this issue， a DOA estimation method based on sparse Bayesian learning （SBL） was proposed in this paper. Within the SBL framework， joint modeling and precision estimation of phase errors were introduced， with Gamma distribution employed to characterize the uncertainty of phase errors on the RIS. Additionally， the expectation-maximization algorithm was utilized to iteratively optimize the posterior probability， enabling accurate correction of phase errors on the RIS and precise DOA estimation. Moreover， to address the problem of signal coherence， unitary transformation was introduced to restore the rank of covariance matrix， enhancing the algorithm’s decorrelation capability under conditions of high signal coherence. The algorithm’s accuracy was effectively improved in resolving multiple DOAs in dense signal environments. Simulation results demonstrated that the proposed method significantly improved DOA estimation accuracy and angular resolution in scenarios with phase errors and highly coherent signals. |
| Key words: direction of arrival （DOA） estimation reconfigurable intelligent surface （RIS） phase error sparse Bayesian learning （SBL） coherent signal |
|
