| Video synthetic aperture radar(ViSAR)has received more and more attention due to its capabilities of high frame rate,high resolution and dynamic scene monitoring which conforms to the human visual system.The terahertz ViSAR has low latency characteristics besides the advantages of ViSAR,which is the best choice for some real-time monitoring applications.With the development of terahertz devices,terahertz ViSAR is emerging.Therefore,it is of great significance to research the structure and imaging algorithms of terahertz ViSAR systems.The key to realizing the terahertz ViSAR system is how to combine ViSAR technology with terahertz technology.Based on this,the key technologies are researched in this dissertation,which contains terahertz ViSAR system structure,high frame rate imaging,motion compensation and moving target detection.The main work and contributions of this dissertation are given as follows:1.For the application requirements of high frame rate,high resolution and low delay,the key parameters of the ViSAR system are demonstrated and the structure design of the terahertz ViSAR system is conducted.This scheme selects the terahertz band as the working frequency,taking into account the requirement of high resolution and low delay.For the problem of the transmission power of current terahertz devices,the scheme that the transmitter is constituted by a solid-state source and a folded traveling wave tube is proposed.In this structure,the radar system can generate a broadband high-power terahertz signal,which provides guarantee for high frame rate imaging of ViSAR system.Aiming at the problem of bandwidth loss for long distance target,a single-source dual-sweep system structure is proposed,which simplifies the system structure while avoiding the loss of system bandwidth.2.To solve the problems of geometric distortion and image defocus in the traditional PFA imaging algorithm,research of the ViSAR real-time imaging algorithm is carried out,a terahertz ViSAR imaging algorithm based on 2-D CZT is proposed.Considering the characteristic that the azimuth accumulation angle is small for ViSAR,the algorithm utilizes CZT to accomplish the wavenumber domain spectral resampling,which reduces the complexity of the algorithm.At the same time,the algorithm solves the problem of geometric distortion and image defocusing based on a more accurate imaging model.And different imaging optimization methods are proposed for different imaging scenarios,which further improve the efficiency of the algorithm while ensuring imaging accuracy.3.Aiming at the motion error of terahertz ViSAR in actual imaging process,a motion error compensation algorithm based on height term removing is proposed to solve the problem of difficulty to compensate for the space-variant error.Utilizing the feature that the height term is not coupled with the target position,the height term is removed from the distance formula and the 2-D space-variant error is simplified to 1-D.Moreover,the residual phase error is compensated by the 1-D scene division,and the quantitative criterion of scene division is given.Based on this quantitative criterion,the compensation algorithm efficiency is improved,while ensuring the imaging quality.4.Aiming at the difficulty of moving target detection and location for the traditional single-transmit and single-receive radar system,moving target detection and location method based on the ViSAR shadow is proposed,which realizes the moving target detection,location,and parameter estimation.This method utilizes the characteristics of the terahertz SAR image,“target-virtual-shadow-real”,to realize moving target detection through shadows.Moreover,utilizing the characteristics of terahertz ViSAR image frame,“adjacent similarity”,a moving target velocity estimation method based on Radon transform is proposed to realize the real-time continuous estimation of the moving target velocity of terahertz ViSAR. |
PDF Full Text Request