基于功率谱密度的盲信号分离

Blind signals separation based on the power spectral density

提出一种新的基于功率谱密度的盲信号分离方法。该方法在传感器数与源数关系不明确情况下,且源信号相互独立时,直接通过混合信号功率谱密度函数比值求解混合阵,并通过混合阵判定观测信号是完备混合、超定混合还是欠定混合,由此进一步分离信号。理论分析、仿真数据证明了算法的有效性。同时还仿真了噪声对分离性能的影响。

A new methods for blind signals separation is proposed. In the case that the relations between the number of sensors and sources are unknown and the source signals are independent, via the ratio of the power spectral density of the observation signals, the mixture matrix is obtained. According to the matrix, it can be assured that the observation signal belongs to one of three mixtures, namely the complete mixture, over-determined mixture and under-determined mixture, and the sources can be separated from the observation signals. The effectivity of the proposed method is verified by theory analysis, simulation data. The influence of noise on separation performance is also simulated.