利用Lissajous图形的雷达辐射源DOA与极化参数联合估计方法
李金朋 , 赵闯 , 胡德秀 , 陈新国
信息工程大学学报 ›› 2025, Vol. 26 ›› Issue (04) : 415 -422.
利用Lissajous图形的雷达辐射源DOA与极化参数联合估计方法
Joint Estimation Method of Radar Emitter DOA and Polarization Parameters Based on Lissajous Figure
稀疏压缩感知类算法无法对相干信号进行波达方向(DOA)估计,极化参数精度估计受幅相误差影响严重。传统算法拘束研究在一维信号层面,对上述问题解决效果不佳,所以尝试利用Lissajous图形获取更高维度的信息。首先,分析信号在空间的电场矢量形成的Lissajous图形与DOA空间关系,结合离网压缩感知(OGJS)算法进行DOA参数估计,改善原算法因灵活度拓展带来的精度损失,信噪比为5 dB时DOA估计绝对平均误差由3.83°降至0.27°;然后,利用调参变分自编码器(β-VAE)对灰度图形式的Lissajous图形仿真数据集进行学习,实现对扰动数据的重构,消除幅相误差。对比卷积自编码器网络处理一维信号,实现信噪比为10 dB、极化相差分辨率在1.5°时,小样本100轮学习训练下,匹配率由80.4%提升至99.2%。
Sparse compressed sensing algorithms cannot estimate the direction of arrival (DOA) of coherent signals, and the accuracy estimation of polarization parameter is greatly affected by amplitude and phase errors. Traditional algorithms, which are mostly restricted to the research of one-dimensional signals, fail to effectively solve the above problems. Lissajous figure is used to obtain higher dimensional information. Firstly, the spatial relationship between the Lissajous figure formed by the electric field vector of the signal in space and the DOA is analyzed, and the off-grid joint-sparse (OGJS) algorithm is used for DOA parameter estimation, which reduces the accuracy loss caused by the flexibility expansion of the original algorithm. When the signal-to-noise ratio (SNR) is 5 dB, the absolute average error of DOA estimation is reduced from 3.83° to 0.27°. Then, the β-variational auto-encoder (β-VAE) is employed to learn the simulation dataset of Lissajous figure in the form of grayscale images, so as to reconstruct the perturbed data and eliminate the amplitude and phase errors. Compared with the processing of one-dimensional signals by convolutional autoencoder networks, when the SNR is 10 dB and the polarization difference resolution is 1.5°, after 100 rounds of small sample learning training, the matching rate is improved from 80.4% to 99.2%.
Lissajous图形 / 离网压缩感知 / 变分自编码器 / 联合估计 / 雷达辐射源 / 波达方向
Lissajous figure / OGJS / VAE / joint estimation / radar emitter / DOA
| [1] |
杨永杰.辐射源极化信息的检测和识别技术研究[D].西安:西安电子科技大学,2012:57-68. |
| [2] |
李永祯,苏翔,张燕琴, |
| [3] |
肖雄,刘桐辛,姜春华, |
| [4] |
逯岩斌,陈文东,杨赟秀, |
| [5] |
郭玉华,常青美,余道杰, |
| [6] |
苑清扬.相干辐射源极化-DOA联合估计算法研究[D].哈尔滨:哈尔滨工业大学,2021:17-20. |
| [7] |
郑轶松,陈伯孝,杨明磊.一种基于空域滤波的空间临近相干源角度估计方法[J].电子与信息学报,2016,38(12):3100-3106. |
| [8] |
|
| [9] |
|
| [10] |
肖思棋.压缩稀疏阵列下基于离网压缩感知的DOA估计算法研究[D].哈尔滨:哈尔滨工程大学,2022:23-40. |
| [11] |
王强,李佳,沈毅.压缩感知中确定性测量矩阵构造算法综述[J].电子学报,2013,41(10):2041-2050. |
| [12] |
|
| [13] |
张梓轩,齐子森,许华, |
| [14] |
徐定藩.李萨如图形的相位参量[J].物理与工程,2001,11(5):27-29. |
| [15] |
郑俊达.基于李萨如图形的相位测量及实现[J].电子世界,2012(3):80-81. |
| [16] |
侯青松,郭英,王布宏, |
| [17] |
|
| [18] |
|
| [19] |
|
| [20] |
|
/
| 〈 |
|
〉 |