阵列流形内插方法(AMI)是一种基于波场模型的宽带聚焦方法。它无须角度预估,可以应用于任意已知结构的阵列,但其采样矩阵的截断点数通常选取较大,尤其在均匀线阵列下大大增加了运算量。针对这个问题,提出了一种采样矩阵优化方法,该方法利用贝塞尔函数的性质,重新构造了采样矩阵,使得新的采样矩阵在不损失信号信息的前提下,最大限度地降低运算量,从而有效缩短了运算时间。计算机仿真实验结果表明了该方法只需和阵元个数相同的截断点数就可以有效地实现方位估计,并获得与阵列流形内插方法相当的方位估计性能。
Array manifold interpolation method is a wideband focusing approach based on wavefield modeling, which does not require initial DOA estimates and can be applied to any array with a known arbitrary geometry. But it usually requires a big number of truncation points in the sampling matrix, especially in the uniform linear array, and the computation is increased greatly. To solve this problem,the paper proposed an optimal method using the character of Bessel function and reconstructed the sampling matrix. The new sampling matrix could spend less computation and time without losing the information of sources. At last the simulation results indicate that the proposed method can estimate the DOAs using the same truncation points to the number of array elements effectively and has a considerable performance with array manifold interpolation method.