波束域的信号盲分离方法

Blind signals separation in array beam space

现有的源信号盲分离方法大都是针对阵元输出信号进行的,各种干扰信号和观测噪声的影响使盲分离算法性能退化,甚至失效。为了提高低信噪比情况下的信号盲分离能力,提出一种新的信号盲分离方法,即先对阵元观测信号进行盲波束形成,而后利用波束输出信号实行盲分离。盲波束形成阶段既提高了盲分离输入信号的信噪比,又可降低盲分离模型的阶次,信号盲分离阶段不仅能进一步净化信号,还能分离同一波束内两个以上的源信号。采用多种情况的人工仿真混合信号进行实验,以评价新算法的性能,仿真结果表明新的盲分离方法优于各阶段算法。

The existing methods of blind signals separation are mainly executed for output signals of array. Under the influence of interfering signals and the noises of observation, the blind sources separation （BSS） algorithm＇s performance is deteriorated, even loses its efficacy. A new approach is proposed to improve the power of blind sources separation under the condition of the lower signal-noise ratio. The blind beamforming is implemented firstly with array＇s signals, and then blind separation is performed in beam space. The blind beamforming process can both enhance effectively SNR and decrease the order of models for BSS. The later BSS algorithm can not only purify signals waveform more over but also separate two or more sources in one beam pattern. In order to evaluate the effectiveness of the proposed method, the experiments are carried out under many conditions of synthesized signals. Simulation results illustrate that the new BSS method is better than tradition algorithms.