Research On Motion Compensation Method For Terahertz High Resolution Imaging

Terahertz SAR inherits the advantages of traditional microwave SAR long-distance detection,all-time detection and all-weather detection.It can achieve higher resolution imaging and obtain richer and more detailed information in the imaging scene.It is more widely used in national defense,military affairs,people's livelihood and national strength.However,clear and high-quality SAR images depend on perfect motion conditions,but there are always motion errors inevitably in practice.The motion errors to be considered in terahertz SAR imaging are more complex and require higher accuracy of motion compensation.Therefore,thesis focuses on the problem of motion compensation in terahertz high-resolution imaging.Firstly,starting from the basic principle of terahertz SAR imaging,it is pointed out that the relative motion between radar and target is indispensable and the motion error is inevitable,highlighting the necessity of motion compensation.Based on the discussion of the basic theory and implementation process of de-linear frequency modulation,the imaging principle and process of Range Doppler Algorithm and Polar Format Algorithm applied to strip and circular bunching modes are analyzed.Several important categories of motion error,phase error and motion compensation mode are briefly summarized.Secondly,aiming at the high-frequency vibration error that can be ignored in traditional microwave SAR but needs special treatment in Terahertz SAR imaging,taking the strip mode as an example,based on the derivation and verification of its influence,the mathematical models of single component and multi-component harmonic superposition are established respectively.The single component high-frequency vibration parameter estimation method based on discrete sinusoid frequency modulation and the multi-component parameter estimation method based on sinusoidal frequency modulation Fourier Transform are given.The estimated values are used to construct the phase factor to realize compensation after range compression and before azimuth processing.In addition,considering the compensation of motion error which is difficult to establish model in terahertz strip SAR,a nonparametric compensation method based on minimizing image entropy is proposed in the thesis.The effectiveness of the above methods is verified by simulation experiments.Finally,aiming at the nonsystematic range cell migration phenomenon introduced by interpolation operation in the common PFA algorithm of terahertz circular SAR and the problem of space variability that cannot be ignored,Chirp Z Transform is used instead of interpolation to avoid nonsystematic range cell migration to a great extent,and a solution of space-varying error is given.The one-dimensional division value of scene range is obtained by simplifying the phase error,and then column by column compensation is carried out by using the all-pass system which is yoked with the residual quadratic phase error.The effectiveness of the scheme is verified by simulation results.