Noise Cancellation in the Presence of Correlated Signal and Noise

Abstract : A model for a multichannel array signal processor that performs a noise-cancellation function is specified. The performance of the processor is examined in terms of the output signal-to-total-background-noise ratio. The parameters of interest are the correlation between signal and noise, the number of noise sampling channels, the relative gain of the noise in the signal-plus- noise versus the noise-only channels, and the array input signal-to-noise and interference-to-noise ratios. An important practical implication of the results presented is that, in systems such as adaptive beamformers for arrays, system performance can be seriously degraded if the signal and noise waveforms are correlated over finite time intervals.