Detection of incoherent pulse trains in non-Gaussian clutter

This paper handles the detection of incoherent pulse trains in compound-Gaussian clutter. We show that the generalized likelihood ratio test leads to a detection structure which is independent of the clutter distributional parameters. Remarkably, it also guarantees the constant-false-alarm-rate property with respect to those statistics. An FFT-based implementation of the structure is proposed. The performance assessment shows that the proposed receiver largely outperform the square-law detector and that the loss of its FFT-based implementation is quite negligible