Research On Ground Moving Targets Detection With Airborne SAR

Ground moving targets detection with airborne SAR plays an important role in the martial field. This dissertation focuses on two key techniques, i.e. clutter suppression and motion parameters estimation. The major work is as follows.First, several airborne SAR/GMTI systems are introduced. Then the study status quo of SAR/GMTI is reviewed and the key techniques as well as the major work are also specified. After the modeling of moving targets'echoes, the motion-induced changes of Doppler history are discussed in detail. It's also analyzed and simulated what influences the changes have on the conventional SAR imaging.The next part deals with clutter suppression methods based on single channel, such as spectral-filter, subaperture-clutter-cancellation and FrFT, and ones based on multiple chanels, such as DPCA, ATI. The simulation results are also given before their performances are compared.The methods of motion parameters estimation are then investigated based on time-frequency analysis. Parameters estimation methods applying STFT and Gabor transform are discussed and compared in brief. Besides, the shortcomings of traditional estimation methods based on WHT and FrFT result in the modified methods, which are able to estimate the initial and center Doppler frequency, and to distinguish between multiple targets lying in the same line in the time-frequency plane. The modified methods are proved by computer simulations.In the end, we conduct a preliminary study on detection of targets with nonconstant acceleration. The new Doppler-time formula are derived and its influence on detection and imaging is analyzed. Then a new SAR/GMTI scheme is given on the basis of the above work. It takes advantage of FrFT to suppress clutters, and to obtain Doppler parameters of targets, with simplicity, practicability and high detection probability as its characters.