摘要
提出一种适用于波束形成后数据的改进的高频天波雷达空间谱重构算法.由于不同接收阵元的目标相位叠加在波束形成后的数据中,当缺少独立阵元输出数据时,已有算法对相同波束宽度内的相干目标难以形成可分辨的空间谱.为此,利用伪逆变换处理波束形成后的数据以获得等效阵列数据,通过结合Toeplitz矩阵构造方法和MUSIC算法重新构造了空间谱,从而实现了相干目标的超分辨处理并改善了方位估计精度.仿真分析了所提方法的有效性以及不同的加权窗对其性能的影响.
Digital beamforming（DBF） is used to transform outputs of a huge array into bearing ones at several expectable directions in the HF skywave radar. However, as the target signal phases on the different array elements are mixed in the beamformed data, the resolvable super-resolution spatial spectra of the coherent targets within a beamwidth can not be directly constructed by the existing algorithms any more. Therefore, an improved reconstruction algorithm for these spatial spectra was investigated to solve the underlying problem. Pseudo-inverse processing was performed to transform the beamformed data to the equivalent array data. Then, combined with the Toeplitz construction and the mulitple signal classification （MUSIC） algorithm, the coherent targets were resolved and the azimuth estimates were improved. Since window weighting is usually used in beamforming, different weighting effects on the improved algorithm were also discussed.
