Research On Channel Characteristics And Imaging Algorithm Design For Synthetic Aperture Microwave Radiometer

Synthetic aperture imaging technology has been widely used in the field of microwave detection and imaging in recent years.However,due to the error of the actual detection and imaging system,the measured value of its visibility function is not accurate,which will lead to the unsatisfactory result of inversion imaging.The synthetic aperture microwave imaging technology is studied in this thesis.In order to improve the influence of error on the system and improve the imaging quality,this thesis focuses on the theoretical research and simulation analysis of the error characteristics of the receiving channel,the correction method for error and the synthetic aperture imaging algorithm based on regularization.（1）The basic principle of synthetic aperture microwave radiometer imaging and the mathematical model of inversion imaging are analyzed and introduced.On this basis,two typical imaging algorithms are analyzed in detail,and the ill condition in the actual imaging process of synthetic aperture is sorted out,which lays a foundation for subsequent research.（2）The influence of channel errors from different sources on visibility is analyzed;For two kinds of errors with different action properties of visibility function,two correction methods are studied,namely,external correction method based on artificial source and sub-band division method;The simulation experiments and performance comparison of one-dimensional point source imaging and two-dimensional spread source imaging are carried out by using these two methods.The simulation results show that these two correction methods can achieve good correction of channel amplitude and phase error and fringe elimination error.（3）The regularization based synthetic aperture imaging algorithm is studied.Aiming at the problem that the filtering function of Tikhonov algorithm suitable for small array elements is too smooth,the Gaussian function is used to optimize its filtering factor;By improving the low-pass filtering characteristics of the regularization matrix,a Gaussian filtering Tikhonov method with better anti noise performance is proposed,and the simulation experiment is carried out.The experimental results show that the regularization based imaging algorithm can suppress the channel amplitude and phase error,fringe elimination error and environmental noise;At the same time,Tikhonov algorithm based on Gaussian filter has better anti noise performance than the original Tikhonov regularization algorithm.（4）Aiming at the problem that the calculation of singular value decomposition of direct regularization method is too complex in the case of large array,the traditional Landweber algorithm is introduced into the field of synthetic aperture imaging,which effectively reduces the computational complexity and improves the imaging efficiency of the system.At the same time,aiming at the slow convergence speed of the traditional Landweber iterative method,the iterative acceleration form of Newton Like method is constructed,and the iterative acceleration Landweber algorithm is proposed and simulated.The experimental results show that the iterative accelerated Landweber algorithm improves the convergence speed and computational efficiency of the algorithm while retaining the imaging accuracy of the original algorithm.