基于改进的AR-MUSIC算法实现方位超分辨

Azimuth Super-resolution Based on Improved AR-MUSIC Algorithm

针对高频地波雷达角度分辨率受限于阵列孔径的问题,利用AR模型循环外推法,对单快拍下回波数据进行合理的外推处理,增加数据长度,相当于接收阵元孔径虚拟增大,然后用MUSIC(multiple signal classification)算法进行DOA(direction of arrival)估计。仿真数据实验表明,改进方法比传统的MUSIC测向分辨率和精度都有明显的提高,且运算量不大,满足高频地波雷达的实时或者准实时高分辨测向要求。

Aiming at the problem of azimuth resolution limited to the array aperture in the High Frequency Sur- face Wave Radar（ HFSWR）, we use the AR-model cycle extrapolation method to extrapolate the received data rea- sonably under the single snapshot, so that the length of the data is increased, which is equivalent to a virtual in- crease of received array aperture, then use MUSIC algorithm for DOA estimation. Simulation experiment results show that the method adopted dramatically improves the azimuth resolution and accuracy comparing with the traditional MUSIC method, furthermore, the method is able to meet the needs of real-time or half-real-time of HFSWR azimuth super-resolution for the less calculation.