A comparison of the detection and resolution performance of multiplicative and additive aerial arrays in the presence of noise

This paper discusses the effect of receiver noise on a multiplicative receiving array. A comparison is made between the low signal/noise performance of an array using multiplicative signal processing and the same array using square-law detection of its signal outputs. A theoretical analysis of the detection performance of a multiplicative array in terms of probability distributions is presented. It is shown that the detection performance of a multiplicative array is identical to a signal processing scheme proposed by Marcum. This scheme involved the subtraction of a square-law detected noise-only pulse from every possible signal and noise square-law detected pulse. The loss in radar detection performance of a multiplicative array compared to an array using square-law detection is found to be quite small for radars that have few pulses per target available for video integration. For a radar that integrates as many as 100 pulses per target the loss in effective input signal/noise ratio is of the order of 1 dB. An experimental X-band array and associated circuitry to perform either multiplicative or square-law processing are briefly described and some photographed video responses to both single and double targets are presented. Limited experimental results showing the probability density functions of the output of the multiplier and square-law detector are plotted. An experimental comparison of resolution performance at low signal/noise ratios of the two types of signal processing is also made. The environments where a multiplicative radar system may prove advantageous are suggested.