An adaptive CFAR threshold determination algorithm based on IR-UWB radar

The problem of inaccurate threshold setting on IRUWB (Impulse Radio-Ultra Wide Band) radar echo signals due to small or large subject movements by the conventional CFAR (Constant False Alarm Rate) algorithm is addressed. In this paper, an adaptive CFAR-based threshold determination algorithm is proposed. The acquired IR-UWB radar echo signal is passed through setting three different MTI filters to obtain three sets of CFAR thresholds. These three values are logically operated to derive the final CFAR thresholds for the test target. Next, the original radar echo signal is passed through the operator of the leakage detection rate to obtain the final MDR value of the tested target. The thresholds are determined by combining the noise signal-based thresholds with the target signal-based thresholds according to the designed weights. Adjusting the designed weights allows the final signal thresholds to be determined based on the critical intersection points. The algorithm proposed in this paper can set the thresholds adaptively for different test target states (e.g., stationary, slight motion, and large motion). The experimental results show that the threshold determination algorithm proposed in this paper is effective and easy to implement.