High-Resolution Beamforming Of Ultra-Wideband Signals Based On Digital Delay Compensation And Array Extrapolation

With the development of radar technology,the functional requirements of radar have been changed from target detection and target positioning,to target tracking,target recognition and target imaging.The development trends of radar system are from single base,single antenna,single pulse and narrowband,to multi-base,multi-antenna,phased array,MIMO,synthetic aperture,pulse compression,wideband and UWB.Increasing the array aperture and the bandwidth of signals are the two key techniques to enhance the ability of detection,the accuracy of positioning and the imaging resolution.Thus high-resolution beamforming of UWB signals with large aperture arrays is a key issue in the field of radar imaging.However,the traditional beamforming method based on phase-shift will cause a lower mainlobe to sidelobe ratio,a limited scanning angle range and beam pointing deviation.Based on the traditional study of beamforming,this dissertation focuses on the beamforming of UWB signals based on digital delay compensation and relative techniques.The main results are summarized as follows:1)The digital delay methods based on fractional delay filters are comprehensively studied.The principles and several typical design methods are introduced.The frequency response of these filters and the reconstruction errors of several typical UWB signals utilizing these filters are analyzed.These researches lay a foundation for the follow-up work on the beamforming of UWB signals.2)The beamforming methods based on Lagrange fractional delay filters are studied,meanwhile simulations utilizing UWB linear frequency modulated signals and UWB chaotic FM signal based on Logistic mapping are carried out.The results show that these methods have drawbacks of the low beam gain and the high sidelobes.To overcome these drawbacks,the beamforming based on Hermite fractional delay filters is studied,and the simulation utilizing UWB LFM signal is carried out.A novel beamforming method based on up-sampling combining with the high order Hermite interpolation filter sets has been proposed on the basis of the above schemes.Simulations results using UWB LFM signals show that compared with the Lagrange interpolation method this method has a better recovery of UWB signals.In addition,compared with fractional delay filters,a better real-time performance,and a simpler structure than fractional delay filters are realized.3)The Caratheodory representation is explored for the UWB signals,and a novel fractional delayer utilizes Caratheodory representation and Hermite interpolation has been proposed.Simulation results utilizing typical signals demonstrate that this method could achieve high interpolation accuracy while keeping a simple sampling structure,and this system is suitable to delay UWB signals and to implement beamforming.4)To further improve the spatial resolution of a limited array aperture for UWB signals,four schemes based on extrapolated array have been proposed.The first one is a beamforming scheme using a virtual array created by a two-dimensional AR model combining with Hermite fractional delay filter sets.The second one is the virtual array beamforming scheme based on Caratheodory representation combining with Hermite fractional delay filter sets.The third one is a planar virtual array beamforming scheme utilizing autoregressive vector extrapolation combining fractional filters based on radical basis functions.The fourth one is the high-resolution time-reversal imaging method utilizing Space-Frequency DORT combining an extrapolated array created by two-dimensional autoregressive vector extrapolation.Simulations utilizing UWB LFM signals confirm that these schemes could improve the spatial resolution of the array significantly without increasing the aperture of the actual antennas.