Blind source separation of interfering signals in analog circuits

A blind source separation method is proposed to identify source interfering signals at analog circuits nodes. Based on an instantaneous linear mixing model, independent component analysis algorithm is used for separating interfering signals from observed signals. The influences of energy storage circuit elements such as capacitance are analyzed via numerical and experimental simulations. It is shown that, for circuits in which energy storage circuit elements perform less significant effects, relatively satisfied separation results of interfering signals can be obtained. On the other hand, in cases that energy storage circuit elements have significant effects, the separation results will deteriorate. Moreover, when a known input signal is enclosed in the observed mixtures, the separation results for interfering signals will be obviously improved.