Performance prediction of OS-CFAR for generalized swerling-chi fluctuating targets

We address the performance prediction of the ordered-statistic constant false alarm rate (OS-CFAR) detector for arbitrarily correlated and possibly nonidentically distributed target echoes in the presence of nonhomogeneous background. Accounting for the generalized Swerling-Chi fluctuating model, including the well-known Swerling models as special cases, we develop the analytic expressions of detection probability in terms of converging series in nonhomogeneous background involving clutter edges and possibly isolated point-like interference. At the analysis stage, we study the truncation error of the derived expressions and then evaluate the impacts on detection performance of the target parameters via numerical simulations.