Research Of SAR Real-time Signal Processing Based On FPGA

Synthetic Aperture Radar (SAR) imaging has characteristics of all-weather, day/night, long range and high-resolution, and has many important and unique applications in disaster detection, resource survey, mapping and military affairs, which makes the research of SAR real-time signal processing significant. The digital signal processing with field programmable gate arrays has developed rapidly and has advantages in the volume, speed and stability over the DSP processors. So SAR real-time signal processing with FPGA will be the necessary choose. The dissertation is about the FPGA design of a real-time azimuth digital preprocessing filter and pulse compression.The paper first analyzes the difficulty in the design of the preprocessing filter and then introduces a real-time fast algorithm for the SAR signal preprocessing from the classical FIR. It combines poly-phase filter with de-sampling and results in a simple architecture for the real-time preprocessing filter. And it is unnecessary to accumulate data and realizes in real-time completely. The design has been fully verified by the real airborne SAR flying experiment.The principle of pulse compression and two methods of the pulse compression realization with FPGA are presented in the paper. Compared with the conventional architecture of pulse compression, the first has the advantage of least resource for reusing the FFT IP. And the second is of high speed.Finally, the effect of quantization finite-word length in the real-time SAR azimuth preprocessing is discussed.