MAP-NSF算法的收敛条件和适用范围扩展

为了拓宽最大后验概率-噪声子空间拟合（MAP-NSF）算法在信号到达角估计上的适用范围，针对通道特性失配和非结构误差两种阵列扰动模型，对MAP-NSF算法的信号到达角估计问题进行了讨论。给出了两种误差模型参数化之间的关系，通过分析，得到了在非结构误差模型下MAP-NSF算法的收敛条件，同时证明了在通道特性失配模型下，MPA-NSF算法可适用于有限样本误差与非结构误差与非结构误差不同量级时的情况。数值仿真结果证实了所得到结论的正确性。

Convergent condition and applied range extension of MAP-NSF algorithm

To extend the applied range of MAP-NSF algorithm, the problem of robust estimation of DOAs of spatial signals under gain/phase error and unstructured error were considered. The parameterization relationship between gain/phase error model and unstructured error model was discussed firstly, after theoretical analysis, the convergent condition of MAP-NSF algorithm was derived under unstructured error and we extended the applied range of this algorithm under gain/phase error. The numerical results validate the theoretical analysis.