A fast algorithm for evaluating the false alarm probability of a CFAR detector

Both the traditional Monte Carlo simulation algorithm and the importance sampling algorithm spend enormous time on evaluating the false alarm probability of CFAR detectors due to the sample loss caused by the "point threshold" [1]. In this paper, a fast algorithm for the evaluation of the false alarm probability of adaptive threshold CFAR detectors is presented by weighting the "point threshold" and then the false alarm probability is estimated with the contribution of all the samples. Analysis and its applications [2] show that the proposed algorithm which succeeds in eliminating the sample loss runs much faster than any other one used before.