复杂运动SAR回波模型与成像方法研究

发布时间：2018-12-12 13:13

【摘要】：合成孔径雷达(SAR)能够在所有的照明/天气条件下产生观察场景的高分辨率微波图像。复杂运动平台SAR以战斗机为载体由于受到外界干扰偏离理想运动状态使得运行轨迹为三维曲线,传统成像算法不再适用。因此如何实现复杂运动平台SAR对观察场景高分辨率成像亟待研究。本文建立了复杂运动平台SAR模型,详细分析了复杂运动SAR与理想运动SAR的区别以及适用于运动误差较小和较大两种情况下的复杂运动成像算法。主要内容分为以下三个部分:首先本文从运动误差、相位误差、多普勒参数和方位向分辨四个方面重点分析了复杂运动SAR与理想运动SAR的区别。在运动误差方面得出方位向回波信号为非线性调频信号,如果不运动补偿将导致方位向的非均匀采样。在相位误差方面分析了不同阶数相位误差对回波信号的影响。在方位向分辨率方面对理论方位向分辨率进行了重新定义:方位向回波信号脉冲压缩后主瓣-3dB宽度。在多普勒参数方面分析了使不同运动模型下得复杂运动SAR的多普勒带宽展宽的加速度容限同时分析了复杂运动多普勒历程的空变性,得出复杂运动SAR需采用BP算法对场景进行逐点成像。然后本文介绍了适用于运动误差较小情况下的复杂运动成像算法:基于IMU实测数据的BP算法。首先通过成像指标得到Legendre正交多项式为该算法高阶拟合的最优基函数结论,然后得出了保证该算法不失效的加速度、速度和位置测量误差容限,此算法只适用于小测量误差情况,当测量误差较大时需要采取基于回波数据参数估计的BP算法。最后本文介绍了适用于运动误差较大情况下的复杂运动成像算法:基于反解Legendre系数强点信息相位跟踪的BP算法。研究了信杂比对相位跟踪算法的影响,得出信杂比门限,重点分析了强点个数和几何分布对雷达位置求解的影响,然后通过设置合适的信杂比和强点个数并通过反解Legendre系数得出雷达平台的斜距史,构造BP算法相位补偿因子并应用于BP算法对场景成像,验证了算法的有效性。
[Abstract]:Synthetic Aperture Radar (SAR) (SAR) can generate high resolution microwave images of observation scenes under all illumination / weather conditions. Because the complex motion platform SAR uses fighter plane as the carrier and deviates from the ideal state of motion due to the external interference, the trajectory is 3D curve, so the traditional imaging algorithm is no longer applicable. Therefore, how to realize the complex motion platform SAR for high resolution imaging of observation scene needs to be studied urgently. In this paper, the SAR model of complex motion platform is established, and the difference between complex motion SAR and ideal motion SAR is analyzed in detail, as well as the complex motion imaging algorithm suitable for the case of small and large motion error. The main contents are divided into the following three parts: firstly, this paper analyzes the difference between SAR of complex motion and SAR of ideal motion from four aspects: motion error, phase error, Doppler parameter and azimuth resolution. The result shows that the azimuth echo signal is a nonlinear frequency modulated signal, and if the motion is not compensated, it will lead to the non-uniform sampling in the azimuth direction. In the aspect of phase error, the effect of phase error of different order on echo signal is analyzed. The theoretical azimuth resolution is redefined in the aspect of azimuth resolution: the main lobe-3dB width after pulse compression of azimuth echo signal. In terms of Doppler parameters, the acceleration tolerance of Doppler bandwidth broadening for complex motion SAR under different motion models is analyzed, and the void variation of complex motion Doppler history is also analyzed. It is concluded that SAR with complex motion needs to use BP algorithm to image the scene point by point. Then this paper introduces the complex motion imaging algorithm which is suitable for the case of small motion error: BP algorithm based on the measured data of IMU. The conclusion that Legendre orthogonal polynomial is the optimal basis function for high order fitting of the algorithm is obtained by imaging index, and then the tolerance of acceleration, velocity and position measurement error to ensure that the algorithm does not fail is obtained. This algorithm is only suitable for small measurement errors. When the measurement error is large, the BP algorithm based on echo data parameter estimation should be adopted. Finally, this paper introduces a complex motion imaging algorithm suitable for large motion errors: a BP algorithm based on phase tracking of strong point information based on inverse solution of Legendre coefficients. The influence of signal-clutter ratio on phase tracking algorithm is studied, and the threshold of signal-to-clutter ratio is obtained. The influence of the number of strong points and geometric distribution on the radar position is analyzed. Then by setting the appropriate signal to clutter ratio and the number of strong points and inverse solving the Legendre coefficient to obtain the oblique distance history of the radar platform, the phase compensation factor of the BP algorithm is constructed and applied to the scene imaging of the BP algorithm, which verifies the effectiveness of the algorithm.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN957.52

【参考文献】