Research On Technologies Of Imaging And Moving Target Detection Based On Compressive Sensing For Spaceborne SAR

Synthetic aperture radar(SAR), as a kind of high-resolution active microwavesensor, not only can observe the earth all-time and whole weather without beingaffected by light and bad weather,but also can effectively identify camouflage andpenetrate the coverings. These advantages make SAR have important applications inboth military and civilian fields. With more demand for higher performanceapplications of SAR, such as imaging in high-resolution and large-bandwidth sceneand high-precision moving target detection, SAR echo processing based on traditionalNyquist sampling brings a huge burden on the system’s hardware device of A/Dconversion, data transmission and storage, etc. But it is expected to solve theseproblems of too large amount of data and too fast of sampling, while we can apply thecompressive sensing (CS) theory, which can break through the traditional Nyquistsampling, to the high-resolution SAR system.Based on the above background, this dissertation mainly aims at the transmittedwaveform design for distributed spaceborne SAR and2-D imaging method andmoving target detection of spaceborne SAR based on compressive sensing. All thesewill be expounded in the following chapters.Chapter one is an introduction of the background knowledge and present situationof SAR, SAR imaging, moving target detection, compressive sensing and theapplication of CS on radar, and then we give the contributions and structuralarrangements of this dissertation.In the second chapter, we firstly analyzes SAR echo and RD imaging algorithm,then we propose a type of composite transmit signal based on linear frequencymodulation and chaotic coding signal. By contrasting the range compression effect,we choose a proper chaotic random sequence, and then we perform simulations indifferent sequence lengths and different satellite number, and verify the effectivenessof the new waveform.In the third chapter, we analyze the2-D sparsity of the radar echo by using CSand Orthogonal Matching Pursuit. The result shows that the SAR echo has sparserepresentation in both cross-range and slant-range. Then we select a suitable under-sampled matrix and construct corresponding reconstructed matrix. At last weget the2-D reconstructed imaging by using CS methods. The simulation results verifythe effectiveness of the imaging method in serious under-sample conditions, and thismethod has better performance than traditional RD imaging.In chapter four, we apply CS theory into azimuth parameter estimation inDisplaced Phase Center Antenna method. While using matched filter in range forpulse compressing, we can find the moving targets and estimate the motionparameters in range. In azimuth, we construct a matching step variably reconstructionmatrix according to chapter three’ imaging method, and then we can search the speedin azimuth of the target step-by-step. Simulation results show the effectiveness of thismethod for moving target motion parameter estimation in the under-samplingconditions.