基于累量域的近场源三维参数联合估计算法

提出了一种无须参数配对的近场源三维参数联合估计算法。通过对阵列输出信号的四阶累积量构造矩阵进行一定的变换构造处理得到一个新的矩阵，仅需一次特征分解并利用其相应的特征值及特征向量就能估计出距离、方位、频率等三维参数。该算法无须谱峰搜索，适于任意高斯噪声环境，参数估计结果能自动配对。并且该算法无须利用阵列的中心对称结构，因而有效地避免了阵列孔径损失。最后，计算机仿真结果证实了该方法的有效性。

Joint estimative method of three-dimensional parameters for near-field sources based on cumulant

A new joint estimative method of three-dimensional parameters for near-field sources without pairing parameters is presented. A new matrix is obtained from the transformation of the constructed four-order cumulant matrix of array output signals. The parameters are directly calculated by the eigenvalues and eigenvactors of the new matrix and just once eigen-decomposition. It does not requre searching spectral peak and can be used to arbitrary Gaussian noise environment, the result of the parameters estimation can achieved automatic pairing. The loss of array aperture is efficiently avoided because the method needs not to use the center of symmetry of the array stracture. Finally, the simulation results verify the effectiveness of the proposed method.