A statistical summary of the standard clutter models is given - Gamma, Inverse Gaussian, Log-normal, Weibull - and it is shown how mixtures of these distributions can be used to represent measured clutter data. This is then related to the problem of detecting a target embedded in clutter and it is shown how this type of clutter characterization can be used to construct a general CFAR detection method. This involves an adaptive adjustment of the detection threshold based on measured characteristics of the clutter. The procedure will maintain a constant false alarm rate as the received clutter data varies over different mixtures of distribution types.
No takes yet. Share an insight, caveat, or question.
Jack M. Miller (1984) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: