The resolution limit of a variable baseline synthetic aperture antenna

The resolution capabilities of a variable baseline correlation interferometer are analyzed using a maximum likelihood estimator algorithm for the output data. Optimum estimates (in the Cramer-Rao sense) for two sources of both equal and unequal strengths, for different levels of receiver-background noise, and for finite integration (averaging) times are obtained. A likelihood probability p_{n}(n = 0, 1, 2, 3, ...) is also obtained from the measured data, and the resolution limit is defined to be that for which the probabilities p_{1} and p_{2} are equal (on the average). It is shown by a computer simulation that the average resolution limit improves (decreases) with increased integration time, exceeding the traditional Rayleigh limit by an arbitrary amount even for quite unequal source strength ratios and large levels of receiver-background noise.