Abstract: Since micro-Doppler phenomenon is observed in radar target detection, micro-motion, as one of significant target characteristics with promising applications in radar target feature extraction and recognition, incur increasing attentions and research interests. The backscatter of target with micro-motions contains fine micro-Doppler features, which is able to determine some unique properties of the target. As the same time, sufficiently many target individual features included in the time-frequency diagram (TFD), which is become an oncoming and useful feature. In this paper , cadence frequency diagram (CFD) feature extraction, which is based on selecting the strongest parts of a cadence-frequency diagram is proposed and studied. Using simulated data we have analyzed this method, which is sound with good classification results.