Order statistics CFAR for Weibull background

N. Levanon, M. Shor

Research output: Contribution to journalArticlepeer-review


The paper shows how to adopt Rohling's order statistics CFAR algorithm, developed for a Rayleigh background, to the case of a Weibull background with a known shape parameter. When the shape parameter is unknown, Weber and Haykin have proposed a two-parameter algorithm in which the threshold is obtained from two ranked background samples. In the paper the CFAR loss of the Weber-Haykin CFAR algorithm is analysed and compared with another CFAR processor for Weibull clutter suggested by Hansen. It is found that, with the optimal choice of representative ranks, the Weber-Haykin CFAR processor yields a smaller, but still very excessive, detection loss. Finally it is shown that CFAR is maintained when more than two ranked samples are used in setting the threshold, but without improvement in performance.

Original languageEnglish
Pages (from-to)157-162
Number of pages6
JournalIEE Proceedings, Part F: Communications, Radar and Signal Processing
Issue number3
StatePublished - Jun 1990


Dive into the research topics of 'Order statistics CFAR for Weibull background'. Together they form a unique fingerprint.

Cite this