Radar Signal Detection Based on the Slope of Singular Curve

A novel method based on the slope of singular curve (SOSC) is proposed in this paper, which is aiming at the signal detection in radar reconnaissance. The SOSC reflects the component rather than the power of the signal, and it is more sensitive to distinguish the radar pulse signals from background noises. Beside that, detecting the pulse signals by their pulse width is adopted, which can farther reduce the false alarm rate. In a low signal noise ratio (SNR) condition, the SOSC based detection method possesses a good performance. The computer simulation with typical radar pulse signals shows that, given the SNR is -2dB and the false alarm rate is 10-3, the detection probability is near to 90 %.