Two level CFAR Algorithm for Multiple Target Detection

Cell Averaging constant False Alarm (CA-CFAR)achieves optimum probability of detection in homogeneous environment but its performance degrades in presence of multiple targets in the reference window. During operational scenarios, the aircrafts fly close by in formation with almost same speed. This results in mutual target masking in range CFAR resulting in detection loss. In this paper, we are proposing Two level CFAR algorithm. In first level, input data is modified only for very high strength targets. The modified input from first level is used to perform CA-CFAR in second level. This algorithm is evaluated with simulated as well as real data for performance metrics such as detection performance with varying SINR, CFAR loss and computational load in comparison with CA-CFAR.