Signal detection and bearing estimation by square-law and multiplicative array processors

Analysis is made of a "multiplicative" receiving array model which is known to have about one-half the beamwidth of the "additive" or conventional (square-law) array with the same number of elements. The probability density functions for the filter outputs of array models are shown for monochromatic, planewave signals in the presence of narrowband Gaussian noise, for broadside arrays with equal spacing and uniform gains. The number of elements, SNR, source bearing, and interelement noise correlation are treated as parameters of the distribution, and their influences are displayed graphically. Based on these probability density functions for the square-law and the multiplicative array processors, we then examine their performances as signal detectors and as bearing estimators.