UWB radar target recognition based on time-domain bispectrum

Complex targets are irradiated by UWB radar, not only the mirror scattering echoes but also the multiscattering interacting echoes are included in target echoes. These two echoes can not be distinguished by classical frequency spectrum and power spectrum. Time-domain bispectrum features of UWB radar signals that mingled with noise are analyzed, then processing this kind of signal using the method of time-domain bispectrum is experimented. At last, some UWB radar returns with different signal noise ratio are simulated using the method of time-domain bispectrum. Theoretical analysis and the results of simulation show that the method of extraction partial features of UWB radar targets based on time-domain bispectrum is good, and target classification and recognition can be implemented using those features.