无源定位技术通过被动接收辐射源信号来实现定位，具有良好的隐蔽性，因而在目标侦察、紧急救援等多个领域得到广泛应用。现有的合成孔径无源定位技术通过平台的移动形成长的合成孔径，可以提升辐射源无源定位的分辨率和精度。然而，基于匹配滤波方法的合成孔径定位方法针对的是长时间全孔径情况下的信号源定位，当信号持续时间不满足全孔径模型时，定位会出现偏差。为此，论文基于子孔径模型开展了时频分布的合成孔径无源定位方法研究，主要工作如下：

（1）提出了基于分数阶傅里叶变换（Fractional Fourier Transform，FrFT）的合成孔径定位方法。建立了子孔径定位模型，基于此模型对其定位方法进行了推导。通过距离频域聚焦，利用FrFT对方位向信号实现方位聚焦。结合坐标转换关系和图像矫正，获得高精度定位图像。通过实验仿真，将其与合成孔径中最大似然估计（Maximum Likelihood Estimation，MLE）的定位方法进行对比，验证了所提方法的效率更高。

（2）提出了基于Radon-Wigner变换（RWT）的合成孔径定位方法。基于子孔径模型对其定位方法进行了推导。通过距离频率聚焦，利用RWT处理方位向信号，从而获得参数域图像。通过方位聚焦加最小熵的方法估计辐射源的调频率。结合坐标转换关系和图像校正处理，获得高精度定位图像。通过实验仿真，将其与MLE和FrFT定位方法进行对比，在保证计算效率的前提下，实现了比FrFT定位方法更高的定位精度。

（3）提出了基于相干积分三次相位函数（Coherently Integrated Cubic Phase Function，CICPF）的快速合成孔径定位方法。基于子孔径模型对其定位方法进行了推导。通过距离频域聚焦，利用CICPF对方位向信号实现方位聚焦，从而获得参数域图像。最后，通过坐标转换关系和图像矫正，获得高精度的定位图像。通过实验仿真，与上述三种定位方法进行对比，验证了所提方法的计算效率和定位精度更优。在此基础上，通过实测数据进一步验证了三种所提定位方法的有效性。

Passive positioning technology realizes positioning by passively receiving signals from radiation sources, which has good concealment, and thus is widely used in many fields such as target reconnaissance and emergency rescue. The existing synthetic aperture passive localization technology can improve the resolution and accuracy of radiation source passive localization by forming a long synthetic aperture through the movement of the platform. However, the synthetic aperture localization method based on the matched filtering method targets the source localization in the case of long full aperture, and the localization will be deviated when the signal duration does not satisfy the full aperture model. For this reason, the thesis carries out the research on the synthetic aperture passive localization method with time-frequency distribution based on the sub-aperture model, and the main work is as follows:

(1) A synthetic aperture localization method based on Fractional Fourier Transform (FrFT) is proposed. A sub-aperture localization model is established, based on which its localization method is derived. By focusing in the distance frequency domain, the azimuthal focusing of azimuthal signals using FrFT. Combined with the coordinate transformation relation and image correction, a high-precision localization image is obtained. Through experimental simulation, it is compared with the localization method of Maximum Likelihood Estimation (MLE) in synthetic aperture, and the proposed method is verified to be more efficient.

 (2) A synthetic aperture localization method based on Radon-Wigner transform (RWT) is proposed. Its localization method is derived based on the sub-aperture model. By focusing in the distance frequency domain, the azimuthal signal is processed using RWT to obtain a parameter-domain image. The tuning frequency of the radiation source is estimated by azimuthal focusing plus minimum entropy. Combined with the coordinate transformation relation and image correction processing, the high-precision localization image is obtained. Through experimental simulation, it is compared with MLE and FrFT localization methods, and a higher localization accuracy than FrFT localization method is achieved under the premise of guaranteeing computational efficiency

 (3) A fast synthetic aperture localization method based on Coherently Integrated Cubic Phase Function (CICPF) is proposed. Its localization method is derived based on the sub-aperture model. By focusing in the distance frequency domain, the azimuthal focusing of azimuthal signals using CICPF. Finally, high precision localization images are obtained by coordinate transformation relations and image correction. Through experimental simulation and comparison with the above three localization methods, it is verified that the proposed method has better computational efficiency and localization accuracy. On this basis, The effectiveness of the three proposed localization methods was further verified by measured data.