摘要
为了提高传统空间平滑算法对相干信号的估计精度,提出了一种张量域空间平滑多重信号分类(Multiple Signal Classification,MUSIC)解相干算法。首先,利用四元数的正交特性重新构造极化域导向矢量;其次,考虑阵列接收数据固有的多维结构特征,构造三阶张量表示的阵列接收数据模型;接着,借鉴二维空间平滑算法的思想,在张量接收数据模型中构建三阶张量子阵块,再使该子阵块在张量域进行前向空间平滑,继而得到平滑后的张量协方差矩阵;最后,通过高阶奇异值分解(Higher-order Singular Value Decomposition,HOSVD)得到信号子空间,利用降维MUSIC算法对相干信号源的二维波达方向(Direction of Arrival,DOA)进行估计,并根据已经获得的DOA信息求解出相干信号的极化参数。仿真结果表明,在信噪比为0 dB以及快拍数为100的情况下,该算法的估计精度比空间平滑算法提高了约70%,成功分辨概率提高了约89%,且无需进行四维谱峰搜索,降低了算法的复杂度,对相干信号具有更高的估计精度和分辨能力。
In order to improve the estimation accuracy of coherent signals by traditional spatial smoothing algorithms,a tensor-domain spatial smoothing multiple signal classification(MUSIC)decoherence algorithm is proposed.First,the orthogonality of the quaternions is used to reconstruct the steering vector of the polarization domain.Second,a third-order tensor representation of the array received data model is constructed by taking into account the inherent multidimensional structural features of the array received data.Then,using the idea of two-dimensional spatial smoothing algorithm,the third-order tensor subarray block is constructed in the tensor received data model,the subarray block is forward spatially smoothed in the tensor domain,and then the smoothed tensor covariance matrix is obtained.Finally,the signal subspace is obtained by higher-order singular value decomposition(HOSVD).The two-dimensional direction of(DOA)arrival of the coherent signal source is estimated by using the dimensionality reduction MUSIC and algorithm,the polarization parameters of the coherent signal are solved according to the obtained DOA The information.simulation results show that the estimation accuracy of the proposed algorithm is about 70%higher than that of the spatial smoothing algorithm,and the probability of successful resolution is about 89%higher than that of the spatial smoothing algorithm when the signal-to-noise ratio(SNR)is 0 dB and the number of snapshots is 100.Moreover,the algorithm does not require searching for four-dimensional spectral peaks and the complexity is reduced,the estimation accuracy and resolution of the coherent signal is higher.
作者
蓝晓宇
单靖炀
梁明珅
董明
马爽
LAN Xiaoyu;SHAN Jingyang;LIANG Mingshen;DONG Ming;MA Shuang(College of Electronic and Information Engineering,Shenyang Aerospace University,Shenyang 2.Liaoning 110136,China;Provincial Key Laboratory of Aerospace Information Perception and Intelligent Shenyang Processing,110136,China;Beijing Institute of Mechanical and Electrical Engineering,Beijing 100074,China)
出处
《电讯技术》
北大核心
2024年第11期1817-1825,共9页
Telecommunication Engineering
基金
国家青年科学基金(61801308)
航空科学基金(2020Z017054001)
辽宁省教育厅面上项目(LJKMZ20220535)
辽宁省自然科学基金(2022-BS-216)
辽宁省青年科技人才“育苗”项目(JYT2020129)。
关键词
极化-DOA估计
解相干参数估计
空间平滑
张量域
polarization-DOA estimation
decoherence parameter estimation
spatial smoothing
tensor domain
