Abstract:
In this PhD Thesis, we address the problem of lower and upper automatic censoring of unwanted samples from a rank ordered data of reference cells, i.e., bilateral or dual automatic censoring, and target detection with Constant False Censoring and Alarm Rates (CFCAR). Assuming a spatially correlated non-Gaussian clutter with no prior knowledge about the presence or not of any clutter edge and/or interfering targets in the reference window, we propose and analyze the
censoring and detection performances of the Dual Automatic Censoring Best Linear Unbiased (DACBLU) CFCAR mono-pulse processor in homogeneous and heterogeneous Log-normal or Weibull spatially correlated clutter. In doing this, we first introduce existing techniques of generating correlated Normal, Log-normal and Weibull random vectors. The cfcarness property is guaranteed by use of linear biparametric adaptive thresholding. That is, we introduce a logarithmic amplifier, and determine the unknown parameters of the transformed distribution through the Best Linear Unbiased Estimators (BLUEs) known to be faster than the ML (Maximum Likelihood) estimators. The Censoring algorithm starts up by considering the two most left ranked cells and proceeds forward. The selected homogeneous set is used to estimate
the unknown background level. To assess the efficiency of the proposed processor, we compare its performances, first to its corresponding BLU-CFAR, i.e., fixed-point(s) censoring detector, then to detectors based on diverse unilateral automatic censoring techniques found in the literature.
