基于解相干TOFS改进算法研究

Decoherence Technology Based on TOFS Algorithm

针对频域子空间正交性测试(TOFS)算法不能处理宽带相干信号波达方向估计(DOA)的问题,提出两种解决方法:一是将修正Toeplitz协方差矩阵重构技术应用到原TOFS算法中,对各频点协方差矩阵做修正Toeplitz化预处理,恢复协方差矩阵的秩与目标数目相等,该方法在低信噪比估计精度有所提高,但会导致阵列孔径的减小,当信源数目较多时,估计偏差较高;二是提出一种基于虚拟变换的解相干TOFS改进算法,该方法利用各频点协方差矩阵的特定元素构造虚拟阵列接收数据,而后进行虚拟空间平滑以达到解相干的目的,与现有算法相比,该算法没有阵列自由度损失,且在低信噪比下算法DOA估计精度有了较大程度提高。数值仿真结果验证了该算法的可行性。

To address the problem that the test of orthogonality of frequency subspace(TOFS)algorithm cannot handle the direction of arrival(DOA)estimation of wideband coherent signals,two methods are proposed in the article.Firstly,the Toeplitz covariance matrix reconstruction technology is applied to the original TOFS algorithm,and the Toeplitz preprocessing is done on the covariance matrix of each frequency point,so that the rank of the covariance matrix is equal to the number of sources.The proposed method improves the estimation accuracy under low SNR,but it brings the array aperture reduction.With the number of sources increases,the estimation deviation increases.Secondly,an improved decoherence TOFS algorithm based on virtual transformation is proposed.This method uses specific elements of the covariance matrix of each frequency point to construct a virtual array to receive data,and then smooths the virtual space to achieve the purpose of decoherence.Compared with the existing algorithm,the proposed algorithm has no loss of array freedom,and under low signal-to-noise ratio(SNR),the DOA estimation accuracy of the algorithm has been greatly improved.Numerical simulation results verify the feasibility of the proposed algorithm.