Research Of Multi-Input And Multi-Output SAR Imaging Method

As a kind of all-day and all-weather remote sensing detection system,Synthetic Aperture Radar(SAR)has been widely discussed and studied in the global since its birth due to the superior imaging accuracy.Nowadays the high azimuth resolution and wideswath coverage have become important performance metrics in the field of SAR imaging with the continuous upgrading of remote sensing technology.Due to the limitation of the minimum antenna area,the traditional single channel imaging mode cannot take into account both high azimuth resolution and wide swath coverage: the high azimuth resolution needs to increase the pulse repetition frequency to avoid Doppler blur,while the wide-swath coverage needs to reduce the pulse repetition frequency to suppress range blur.As a result,multi-input multi-output SAR imaging technology emerged.Multiple receiving channels in azimuth receive echoes at the same time with reducing the temporal sampling rate at the cost of increasing the spatial sampling rate.Multiple transmitting channels in range radiate pulse signals respectively,which can obtain more spatial sampling to improve the imaging performance.This paper focuses on the key issues in the process of MIMO SAR imaging processing,and the main contents are as follows:1.Author studies the problem of signal reconstruction and motion error correction in MIMO SAR system and sets up the echo and motion error models.The influence of multi-channel reception on the echo signal in time as well as frequency domain is researched carefully and the author expounds the reasons for the appearance of Doppler fuzzy.The azimuth signal reconstruction method is derived in detail and the motion error influence on imaging quality is analyses.What's more the effectiveness of the research above are verified by simulation experiments.2.The problem of echo separation and imaging processing of MIMO SAR system are studied.The multidimensional echo separation model is established,and the echo separation method based on range Doppler beamforming technology is analyzed in detail and verified by simulation experiments.3.A sparse reconstruction method of MIMO SAR based on covariance matrix is proposed.According to the relative geometric relationship between the motion platform and imaging scene,author sets up a convolution model between the target scattering coefficient and the antenna pattern,furthermore the convolution inverse problem is transformed into an optimization problem.In this paper a cyclic minimization algorithm is presented and the sparse estimation is implemented online.An additive factor is introduced in order to prevent the error amplifier during iteration.Different from the traditional pulse compression theory based on Fourier transform,the proposed method achieves good noise smoothing and sidelobe suppression effects,which is verified by simulation experiments.