Imaging-tracking Technology For Hidden Targets Of A Through-the-wall Radar

Through-the-wall imaging radar(TWIR)can detect,locate,track and image the hidden targets by passing an electromagnetic(EM)wave through building walls,which is of great importance in both civilian and military applications,such as the counterterrorism,the urban war,and the disaster rescue.Compared with conventional radar,it is much more challenging to robust track the targets behind walls,since the TWIR often works in complex EM environment resulting in the phase distortion and multipath propagation.In this thesis,we address the problem of target tracking using the TWIR with an ultra-wideband ultra-sparse array.The contributions are listed as follows:1.The image autofocusing estimation algorithm is proposed to derive the unknown wall parameters and the refraction angle approximation algorithm is presented to refocus and reposition the target image with the estimated wall parameters.2.To suppress the multipath ghosts,we present the multiple sub-aperture images multiplication algorithm and the exponential phase coherence factor(EPCF)weighting algorithm.Meanwhile,the analytical expression of the efficacy of the EPCF weighting algorithm is derived.We develop the modified sign coherence factor(SCF)weighting algorithm to suppress the grating lobes of the ultra-wideband ultra-sparse array,which can achieve similar image quality as the EPCF weighting algorithm but use 84 percent less calculated amount.3.The modified mean shift algorithm is developed to track the human targets in image domain,where the two-axis orthogonal projection method is exploited to extract the target template and the greatest similarity of the target image between the current frame and the target template is utilized to update the target position.Combining with an adaptive adjustment strategy of the MTI structure,the modified mean shift algorithm realizes the robust tracking of the human targets behind wall.4.The building layout aided target tracking algorithm is proposed to improve the tracking robustness of multiple targets in different rooms.First,we use the coherent integration method of multiple pulse,the inversion method of first-order multipath and the single-side two-location spotlight imaging method to estimate the building layout.Then,under the estimated building layout,we obtain the high-quality images of all rooms,where each room uses a special time-delay to compensate the wall influence.Finally,based on the formation mechanism of the multipath propagation,the building layout is employed to remove the multipath ghosts.All the aforementioned algorithms have been verified by through-wall-radar experiment data or electromagnetic simulation data.