Simulation Of Raw Signal And Image Of Synthetic Aperture Radar

Synthetic aperture radar (Synthetic Aperture Radar: SAR) has been widely used in military and civilian fields. Techniques of synthetic aperture radar raw signal simulation and image simulation play an important role for the SAR system design, validation, SAR imaging algorithm testing and automatic interpretation of SAR images, etc. These techniques are deeply studied in this paper. Two aspects of SAR simulation studies are carried out in this paper: the SAR raw signal simulation and image simulation of the typical terrain, as well as SAR image simulation of the typical targets. The framework for the SAR raw signal simulation and image simulation of the typical terrain has been established. The SAR image simulation of typical targets has completed. The major work and contributions are listed below:1) A new anisotropic terrain generation method based on combined fractal is proposed in this paper, which accords with the natural terrain features. Two kinds of shape-controlled fractal terrain generation methods are proposed in this paper, respectively. One method is based on the initial control points and the other method is based on the combined surface. By utilizing these methods, the generation of typical terrain based on fractal geometry is achieved.2) A complete set of the SAR raw signal and image simulation for the typical terrain is established in this paper, which includes the terrain generation, rough surface scattering calculations, SAR raw signal simulation and SAR imaging algorithm. The SAR image modeling of the typical terrain is completed. The raw signal simulation error between frequency domain method and time domain method is analyzed in detail in this paper.3) An improved SAR raw signal simualtion method in the case of non-ideal tracjectory is proposed in this paper. In the case of the assumptions of the small azimuth beam angle, the formulas are re-derived in this paper and the azimuth inverse Fourier transform is eliminated during the integration of the range. Some calculation parts in the origianl algorithms are merged. By these two methods, the simulation speed is accelerated. In the case of the assumptions of the small azimuth beam angle and middle trajectory deviation velocity, the SAR system transfer function is separated into the range-dependent function and azimuth-dependent function in this paper. The computation amount of the range integration is reduced and the simulation speed is accelerated.4) An efficiency raw signal simulator in the frequency-domain of bistatic SAR is proposed in this paper, in the case of parallel flights and translational invariance conditions: Through the analysis of the range history of the bistatic SAR system in the side-looking mode with ideal tracjectory, the side-looking mode with non-ideal tracjectory as well as the squint-looking with ideal tracjectory, several equivalent formulas between the bistatic SAR and monostatic SAR are derived in this paper. Based on these formulas,the fast raw signal simulation methed in the frequency-domain of the monostatic SAR is extened into the case of bistatic SAR. The simulation results prove validity of this equivalent method.5) The SAR image simulation method for the typical targets based on pixel method is proposed in this paper. Firstly, the complex electromagnetic scattering coefficience of each pixel is calculated and then the scattering coefficience is converted to the corresponding SAR images through the vector superposition method. The simulation results prove that this method is faster and effective. Also the turntable radar image simulation method for the typical target is given in this paper. By using the frequency step method, the raw signal of the typical targets in the turntable mode is simulated and the turntable radar images is obtained by the corresponding imaging algorithm.