| Synthetic aperture radar(SAR) is a kind of high-resolution radar which can achieve high resolution respectively by pulse compression techonology in range direction and doppler frequency analysis techonogoly in azimuth direction. Therefore, it can acquire the 2-D image of one scene. Compared to optical imaging, SAR can work under all weather, day and night, which is indenpedent of many natural conditions such as night, cloud and mist etc. Furthermore, SAR can penetrate the earth’s surface and vegetation under certain frequency, so it can obtain the target information under some coverings which makes SAR has broad application foreground in civil fields such as geological exploration, topographic mapping etc. and in military fields such as taget recognition, battle reconnaissance etc.The combination of the SAR technology and frequency modulation continuous wave(FMCW) technology provides conditions for the development of the small volume, low cost, high reliability and high resolution imaging sensor. The highly miniaturized FMCW SAR system really satisfies the requirement of mini umanned aviation vehicle(MUAV) whose payload is strictly limited. Therefore, FMCW SAR plays an important role in both military and civil fields, and more and more research organizations show great interests in it. But compared to pulse SAR, FMCW SAR transmits frequency modulated signal continuously, so the long signal length deteriorates the distortion of received signal, and the motion of radar platform within a pulse repetition interval can not be ignored, which makes the ’stop-go-stop’ assumption(The change of instant slant-range between target and radar during a pulse repetition interval is ignored) which is derived in traditional pulse SAR imaging algorithm no longer valid. Thus, the two differ in echo model and imaging processing. Therefore, the existing imaging algorithm under pulse SAR cannot be directly used to the FMCW SAR, and it must be improved by combining the characteristics of FMCW SAR echo signal. Based on FMCW SAR imaging processing and system implementation, this thesis mainly complete the following work.1) The characteristics of FMCW SAR echo signal and the influence on echo signal by Dechirp process are analysed. The difference between FMCW SAR echo model and pulse SAR echo model is analysed in detail. And the instant frequency spectrum of FMCW SAR echo is analysed, then the echo model is built. And the influence on range signal from doppler distortion is analysed, then whether the influence can be ignored is analysed, and the conditions of ’stop-go-stop’ assumption are pointed out. The analysis is validated by simulating FMCW SAR echo.2) The FMCW SAR imaging algorithm is studied by improving the traditional pulse SAR imaging algorithm. In pulse SAR imaging algorithm, the range doppler(RD) algorithm and the chirp scaling(CS) algorithm are mostly commonly used, however, neither of the two algorithms take the change of slant-range form continuous motion of radar within the pulse transmitting into consideration, which certainly leads to the doppler distortion and the additional range migration. If not corrected, it must cause the spread of range and azimuth image or offset, and thus the system resolution cannot be maintained. In the thesis, the RD algorithm is improved in Chapter 3 for FMCW SAR. In addition, the CS algorithm makes the range migration tracks of different slant-range be the same through phase multiplication, and then corrects the residual range migration in bulk, which solves the problem that the range migration changes with different slant-range and improves the imaging precision as well as the computational efficiency. However, CS algorithm demands that the range signal must have liner frequency modulated characteristic, which is certainly not suitable for FMCW SAR. The principle of frequency scaling(FS) algorithm is similar to CS algorithm, which elimates the difference in range migration tracks of different slant-ranges through phase multiplication and is suitable for FMCW SAR echo received by Dechirp. However, the tranditional FS algorithm also doesn’t take the motion of radar platform within a pulse repetition interval into consideration, which undoubtedly deteriorates the final imaging quality. In Charpter 4, the tranditional FS algorithm is improved for FMCW SAR.3) The polarimetric imaging of FMCW SAR is studied. Polarimetric measurement is an important development trend of SAR system. So the study to acquisition, description and dipose of polarimetric information becomes an important development trend of polarimetric imaging. In this thesis, the basic theroy of polarimetric imaging is firstly elaborated, and the research emphasis is the polarimetric decomposition theory. The common coherent and incoherent decompositions are are described, and their characteristics are analysed and compared. Because the model based decomposition is directly related to the appreciable physical scattering mechanisms and the decomposition result is visualized and understood, so it has become the research hotspot of polarimetric imaging at home and abroad. Based on these, in this thesis, the model based decomposition approach is studied in depth. This thesis points out the deficiencies in the existing model-based decomposition approaches and presents a more adaptive model-based decomposition for polarimetric SAR data interpretation, and an optimization measure is then proposed to best fit data into models. The experiment results show great improvements using the AIRSAR and E-SAR data sets.4) The system design and implementation are finished. And the feasibility is validated through flight test. The system uses two separated antennas to improve the isolation, and the antenna adopts the way of lens horn to further reduce the weight of the system and decrease the size of the antenna.At last, the full work of this thesis is summaried and the prospect is presented. Problems in FMCW SAR imaging and system implementation which are solved in this thesis are summaried, and the remaining problems to be solved next are also pointed out.
|
PDF Full Text Request