提高二维DOA估计分辨率的改进MUSIC算法

Modified MUSIC algorithm for improving the resolution of 2-D direction of arrival estimation

针对经典二维多重信号分类(Multiple Signal Classification,MUSIC)算法在低信噪比和小快拍数情况下,分辨率受阵列孔径限制的问题,提出了一种改进的基于MUSIC算法的二维测向算法.该方法利用MUSIC谱函数极大值点处对方位角和仰角的二阶偏导数小于零的特性,通过对方位角和仰角求二阶偏导,构造了新的空间谱函数.对新的空间谱函数进行谱峰搜索,其负向谱峰所对应的角度就是目标的波达方向(Direction Of Arrival,DOA)估计.理论分析和仿真结果表明,在低信噪比、小快拍数下,该方法对相近信源有更高的角度分辨率和更低的均方根误差,并且可适用于任何阵型.

The resolution of the classical two-dimensional（2-D） Multiple Signal Classification（MUSIC） algorithm is limited by the certain array aperture under the contexts of low signal to noise ratio（SNR） and finite snapshots.Focusing on this problem,a modified 2-D direction of arrival（DOA） method based on the MUSIC algorithm is proposed.By calculating the second partial differentiation for the azimuth and elevation of the MUSIC spectrum,some negative peaks are shown around the original DOAs while MUISC spectrum does not make peaks there.The performance of the proposed method is evaluated for separating closely spaced sources in difficult contexts,for instance,a low SNR and a small number of snapshots in comparison with 2-D MUSIC method through computer simulations.Furthermore,the new approach is suitable to any array shape.