Research On The High Resolution Airborne SAR Real Time Processor

High and super-high resolution airborne Synthetic Aperture Radar (SAR) is an important research orientation. Currently, the advanced countries are trying their best to develop high-resolution airborne SAR systems. The research work of this paper goes with the development of real time SAR processor in our research group. A study of the special characteristics in the real time SAR processor is presented.I/Q error is the first problem faced to the SAR signal. I/Q amplitude and phase error can create an image response at the negative frequency, which cause serious disturbance to signal detection. The SAR echo studied here, the imaging algorithm is adopted to compress pulses, which is a two dimension matched filter. When the I/Q error is constant in an aperture, it will do nothing to the signal after the pulse compression. Considering the real time computation, in this SAR processor the I/Q correction is not considered.Motion compensation is an inseparable part to SAR. Combined with the real time SAR processor, a motion compensation realization is introduced in this paper. That includes the first order motion compensation, which is important and necessary, the second order motion compensation, whose phase is miner and has a large computation, in real time SAR processor, it is not be considered, autofocus is obtained by plug in. Furthermore, autofocus only can solve defocus problem, this paper introduces an idea to correct the image geometric distortion, this method uses the result of PGA, by which a phase multiplication corrects the geometry distortion. When the motion error of phase center passes over a range bin, resample should be taken into consideration, the resample inaccuracy within one range bin does not have impact on azimuth focus, but to the signal amplitude.This real time SAR processor has side-looking and squint modes. Two approaches are proposed to use the side-looking raw data to squint imaging. The first one is the direct transformation, which uses the data with a wide beam angle. The maximum of the E-SAR raw data squint angle is 7°. The second one is the indirect transformation, which is carried out by adding a platform velocity along slant range according to the required squint angle. Then the squint data is determined by the angle between the new forward velocity and LOS direction. The drift angle is less than 15°, the indirect transformation is showed acceptable.Squint imaging has some specialties, which are different under squint condition, compared four mainly squint imaging algorithms, a new extend Chirp-scaling algorithm is presented to eliminate the cubic phase error due to the slant range expanding to the cubic term under high squint angle.