Abstract:The resolution of close targets in a tested unit has always been a hot research spot in various signal processing fields. The Rayleigh limit, which is based on the correlation analysis or beamforming, is hard to reflect the stochastic influence as the noise existing in the actual system. Therefore, the concept of statistical resolution has been proposed. There are two definitions of statistical resolution limit (SRL)—— one is based on the Cramér–Rao lower bound and the other is the binary hypothesis testing theory. In this paper, the current research of the latter has been concluded, and a general solution has been obtained. The binary hypothesis test is applied to model the resolution problem, which is solved by resorting to a linear model after the Taylor expansion and approximation, or with the conclusion of the asymptotic distribution under the condition of a large number snaps. Based on those tools, the relationship between SRL and other variables, such as waveform, SNR etc., has been obtained. Finally, some future research suggestions are also presented.