为进行高分辨到达角(DOA)估计的同时避免稀疏类算法的不足，提出了协方差拟合旋转不变子空间信号参数估计(ESPRIT)算法.首先将协方差拟合准则转换成半正定规划问题，利用凸优化进行求解，得到更接近理论值的信号协方差矩阵；然后对估计的信号协方差矩阵进行特征分解，利用信号子空间和噪声子空间特征值的差异估计信源个数；最后利用子空间旋转不变性反解出未知DOA.仿真实验从DOA估计精度、分辨率等方面验证了该算法的有效性，较传统ESPRIT算法具有更高的DOA估计分辨率并且受相干信源影响小；与稀疏类算法相比，不依赖先验信息以及避免了网格失配问题.
Spare representation based direction of arrival (DOA) estimation algorithms have the merits of high resolution and good adaptability to coherent signals, while they suffer from the shortages of depending on prior information and offgrid problem. In order to get high DOA estimation resolution and avoid the shortcomings of spare representation algorithms, covariance fitting estimation of signal parameters by rotational invariance technique (ESPRIT) algorithm is proposed in this paper. Firstly, covariance fitting criterion is converted into a semidefinite programming problem solved by convex optimization. As a result, a signal covariance matrix which is closer to theoretical value is obtained. Then, eigen decomposition of the estimated signal covariance matrix is conducted and the number of signals is estimated through the differences between eigenvalues of signal subspace and noise subspace. Finally, rotational invariance technique of signal subspace is used to estimate DOA. Simulation experiments have proved the validity of the method. Compared with traditional ESPRIT algorithm, the proposed algorithm has higher DOA estimation resolution and is slightly affected by coherent signals. It is also superior to spare representation algorithms for independence of prior information and avoidance of the offgrid problem.
［1］STOICA P, BABU P, LI J. New method of sparse parameter estimation in separable models and its use for spectral analysis of irregularly sampled data［J］. IEEE Transactions on Signal Processing, 2011, 59(1): 3547.
［2］毛琳琳，张群飞，黄建国，等. 基于互相关协方差矩阵的改进多重信号分类高分辨波达方位估计方法［J］. 电子与信息学报，2015, 37(8): 18861891.
MAO Linlin, ZHANG Qunfei, HUANG Jianguo, et al. Improved multiple signal classification algorithm for direction of arrival estimation based on covariance matrix of crosscorrelation［J］. Journal of Electronics & Information Technology, 2015, 37(8): 18861891.
［3］CAO R, WANG C, ZHANG X. Twodimensional direction of arrival estimation using generalized ESPRIT algorithm with nonuniform Lshaped array［J］. Wireless Personal Communications， 2015, 84(1): 321339.
［4］LI Y, CHI Y. OfftheGrid line spectrum denoising and estimation with multiple measurement vectors［J］. IEEE Transactions on Signal Processing， 2015, 64(5): 12571269.
［5］CHI Y, SCHARF L L, PEZESHKI A, et al. Sensitivity to basis mismatch in compressed sensing［J］. IEEE Transactions on Signal Processing, 2011, 59(5): 21822196.
［6］ZHANG Y, YE Z, XU X, et al. Offgrid DOA estimation using array covariance matrix and blocksparse Bayesian learning［J］. Signal Processing, 2014, 98(13): 197201.
［7］YANG Z, XIE L, ZHANG C. OffGrid direction of arrival estimation using sparse Bayesian inference［J］. IEEE Transactions on Signal Processing, 2013, 61(1): 3843.
［8］YANG Z, XIE L. Enhancing sparsity and resolution via reweighted atomic norm minimization［J］. IEEE Transactions on Signal Processing, 2015, 64(4): 9951006.
［9］BHASKAR B N, TANG G, RECHT B. Atomic norm denoising with applications to line spectral estimation［J］. IEEE Transactions on Signal Processing, 2013, 61(23): 59875999.
［10］BLU T, DRAGOTTI P L, VETTERLI M, et al. Sparse sampling of signal innovations［J］. IEEE Signal Processing Magazine, 2008, 25(2): 3140.
［11］YANG Z, XIE L. On gridless sparse methods for line spectral estimation from complete and incomplete data［J］. IEEE Transactions on Signal Processing, 2015, 63(12): 31393153.
［12］YANG Z, XIE L, ZHANG C. A discretizationfree sparse and parametric approach for linear array signal processing［J］. IEEE Transactions on Signal Processing, 2014, 62(19): 49594973.
［13］OTTERSTEN B, STOICA P, ROY R. Covariance matching estimation techniques for array signal processing applications［J］. Digital Signal Processing, 1988, 8(3): 185210.
［14］STOICA P, BABU P, LI J. SPICE: A sparse covariancebased estimation method for array processing［J］. IEEE Transactions on Signal Processing, 2011, 59(2): 629639.
［15］RAO B D, HARI K V S. Performance analysis of ESPRIT and TAM in determining the direction of arrival of plane waves in noise［J］. IEEE Transactions on Signal Processing, 1989, 37(12): 19901995.
［16］HE Z, CICHOCKI A, XIE S, et al. Detecting the number of clusters in nway probabilistic clustering［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(11): 20062021.
［17］SAHOO S K, MAKUR A. Signal recovery from random measurements via extended orthogonal matching pursuit［J］. IEEE Transactions on Signal Processing, 2015, 63(10): 25722582.
［18］JING Y, FENG N, SHEN Y. Modified MUSIC estimation for correlated signals with compressive sampling arrays［J］. Journal of Systems Engineering and Electronics, 2014, 25(5): 755759.
