基于頻段能量归ー化处理的自适应波束形成方法

给出了一种基于频段能量归一化处理的宽带最小方差无失真响应算法。该算法首先将宽带信号的各窄带快拍向量进行能量归一化处理，然后计算窄带输出功率，最后通过对各窄带输出结果进行平均或求和得到最终空间能量谱。它不需要阵列噪声和目标信号的任何先验信息，只要阵列噪声的频谱幅度在频段间的起伏较为剧烈，而目标信号的频谱幅度起伏较为平缓，该方法就能取得较常规处理方法更优的信号检测性能。利用最优信号检测理论证明了该算法的有效性，利用仿真和实际数据进一步验证了算法的性能。仿真结果表明：文中算法较常规处理方法的处理增益提高约4 dB.

Adaptive Beamformer Based on the Frequency Energy Normalization

An algorithm of broadband minimum variance distortionless response (MVDR) based on the frequency energy normalization was proposed. It is the algorithm flow that every narrowband frequen?cy component of the broadband signal is normalized by the total narrowband energy of all array ele?ments, the narrowband output power is calculated by MVDR, final spatial energy spectrum can be ob?tained by averaging or summing all results of every narrowband frequency bin. Any prior-information about the noise or the signal is unnecessary for the proposed algorithm ? The processing gain of the pro?posed algorithm compared to the conventional broadband MVDR can be obtained as long as the ampli?tude fluctuation of the array noise frequency spectrum is severer than that of the target signal. Based on the optimal signal detection theory, the validity of the proposed algorithm was validated in theory. simulation and real data are used to validate the effectiveness of the algorithm. The analysis results show that the processing gain can be improved by 4 dB with the proposed algorithm compared with the conventional one.