Study On The High-Precision Imaging Algorithm Of The L-Band 1-D Synthetic Aperture Microwave Radiometer System

Soil moisture and ocean salinity are the key parameters for studying the characteristics of global water cycle change.In order to achieve the accurate measurements of these two parameters,the research on spaceborne L-band microwave radiometer has been a popular research field in recent years.Compared with other microwave frequency bands,the L-band is widely used for spaceborne radiation measurement of these two target parameters since the L-band brightness temperature of the ground object is more sensitive to soil moisture and ocean salinity.The L-band radiometer systems that have finished on-orbit operation includes: a two-dimensional synthetic aperture microwave radiometer system(ESA SMOS/MIRAS)and a real aperture microwave radiometer(NASA Aquarius,SMAP missions).According to the existing development experience and lessons,it can be known that for the two-dimensional synthetic aperture radiometer MIRAS,although it is easy to meet the spatial resolution required for spaceborne detection,it has the problems of high system complexity and low imaging accuracy.For real aperture radiometers,however,the bottleneck technique lies in spatial resolution.Considering comprehensively,the Lband one-dimensional synthetic aperture microwave radiometer system is an effective tool for remotely measuring ocean salinity and soil moisture,and has a good app lication prospect.With the continuous development of the L-band radiometer,the technical challenge of measuring the ocean salinity and soil moisture is mainly to improve spatial resolution and radiometric accuracy.In order to achieve the spatial resolution that meets the application requirements under spaceborne observation conditions,the one-dimensional synthetic aperture microwave radiometer replaces the large real-aperture antenna in the conventional radiometer by using a sparsely arranged antenna array in the cross-track direction,thereby achieving a better expansion space in spatial resolution.However,for the one-dimensional synthetic aperture radiometer system,there are still serious challenges to improve the accuracy of radiation measurement.Since the difficulty of carrying and on-orbit scanning techniques limits the design size of the antenna,in practical applications,the number of antennas in the spaceborne onedimensional radiometer system is less,which will cause the sampling truncation in the spatial frequency domain,affecting the accuracy of the brightness temperature image reconstruction.In addition,there are inevitably uncertainty errors,side-lobe deterioration,and other ripple error in the antenna pattern data of the radiometer system,which will lead to a decrease in the accuracy of the radiation measurement.Therefore,in order to achieve the accuracy of the radiometer measurement required by the salinity mission,the systematic research on the high-precisionimaging method for the one-dimensional synthetic aperture microwave radiometer is done in this paper.The main research and achievements are as follows:1.For the first time,The theoretical imaging error limiting factors for the high-precision measurement of the one-dimensional interferometric synthetic aperture radiometer system with limited baselines are analyzed from the system perspective.The Gibbs error,image errors caused by aliasing and antenna error are analyzed and the corresponding correction methods were studied.Defining the imaging error of the one-dimensional synthetic aperture microwave radiometer system: there is a non-zero difference between the true image of the target and the image reconstructed after the calibration process.And usually,the reconstructed brightness temperature image can rougthly meet the application requirements.The imaging error studied in this paper is used to quantitatively evaluate the imaging accuracy of the radiometer system.It is found that the influence factors of the imaging error of the one-dimensional synthetic aperture radiometer mainly include two aspects: the baseline design parameters of the antenna array and the antenna pattern data.For the one-dimensional synthetic aperture radiometer with a small number of feeds,there is a truncated sampling in the spatial frequency domain,resulting in significant Gibbs error in the imaging results.Aiming at this problem,this paper firstly proposes and validates the Gibbs error suppression method based on baseline optimization and CLEAN algorithm.In the actual design process of the L-band synthetic aperture radiometer system,limited by the size of the feed,usually the shortest baseline of the radiometer will be greater than the Nyquist sampling limit.Therefore,there is imaging aliasing.Aiming at the imaging error introduced by aliasing in one-dimensional system,the effectiveness of the corresponding suppression algorithm is studied and verified by simulation experiments.In a one-dimensional radiometer antenna array system,the positions of the feeds are different,the distances from the center of the reflector are different,and the size of the feeds are different,which cause the inconsistency of the antenna patterns.Due to the limited machining accuracy of the antenna device and the unstable environment caused during launch and on-orbit operation,there are inevitably has uncertainty errors in the antenna pattern data.The inconsistencies and uncertainty errors in the antenna patterns can introduce the radiometer imaging errors.Therefore,by evaluating the influence of the antenna error on the imaging quality,the antenna array arrangement of the radiometer system can not only be optimized,but the design paremeter of the antennas can also be fed back.In this paper,the radiometer imaging error caused by the sidelobe deterioration of the antenna patterns is analyzed in detail.And the modified G-matrix method is proposed and verified to calibratethe imaging error cased by the antenna sidelobe deterioration.2.Antenna error is one of the main contamination factors in the current radiometer measurements,and the antenna error is outside of the internal calibration term of the radiometer system,and can not be completed by self-calibration.Therefore,the imaging error correction algorithms are studied to calibrate the antenna error,which are an external calibration algorithms of the synthetic aperture microwave radiometer system.Based on the idea of flat target transformation algorithm(FTT),combined with the residual bowl-shape imaging error in the reconstructed image after applying the FTT algorithm,an improved external calibration algorithm is proposed: the reference target transformation algorithm(RTT).The correctness of the RTT calibration algorithm is verified by imaging simulation experiments on the ocean scene,and the calibration effects of FTT and RTT are compared and analyzed.The influence factors of RTT calibration accuracy include: the selection of reference scenes and the modeling accuracy of the brightness temperature image of the reference scene.Combined with the research on the above calibration algorithm,a set of calibration procedures suitable for the one-dimensional synthetic aperture microwave radiometer system is summarized.3.Based on the system design,imaging principle,and calibration algorithm of the synthetic aperture radiometer studied abrove,a one-dimensional radiometer simulator is build and improved.The radiometer simulator mainly includes three modules: the forward brightness temperature generation module,the radiometer system measurement module and the image reconstruction module.The first module realizes the conversion from the meteorological data(sea surface salinity,sea surface temperature,wind speed,water vapor and liquid water and so on)to the brightness temperature data of the target scene.The second module is the simulation of the radiometer observation process,which obtains the visibility function samples from the brightness temperature image.And the third module simulates the reconstruction process of the brightness temperature image.At present,the functions of the end-to-end simulation system include: calculating the orbit positions of the spaceborne radiometer,evaluating the imaging or calibration algorithms,analyzing the influence of the external error sources on the imaging accuracy of the radiometer instrument,optimizing the the antenna feeds arrangement of the one-dimensional synthetic aperture radiometer,and so on.This simulator has already served the designs of the Interferometric Microwave Imager(WCOM/IMI)and Microwave Imager Combined Active and Passive(MICAP).