【作者】 王超；

【导师】 李绍滨；

【作者基本信息】 哈尔滨工业大学 ， 信息与通信工程， 2015， 博士

【摘要】 合成孔径雷达是一种高分辨率微波成像雷达,具有全天候、全天时和远距离的工作特点,被广泛应用于国家安全和国民经济等领域。根据工作方式的不同可分为SAR(Synthetic Aperture Radar)和ISAR(Inverse Synthetic Aperture Radar)两大类。当目标的机动性较小时,某一距离单元内的目标回波信号可近似为线性调频(Linear Frequency Modulated,LFM)信号;而当目标的机动性较强时,目标回波信号应近似为多分量多项式相位信号(Polynominal Phase Signal,PPS信号)。在成像过程中,目标成像质量的高低取决于信号处理方法的优劣。因此,可采用信号处理的方法来提高机动目标ISAR成像的质量。本文主要从信号的参数估计、时频分析和信号分解三个方面处理以下情况的机动目标回波信号模型,分别是:LFM信号模型、三次PPS信号模型、四次PPS信号模型、任意阶次PPS信号模型和调频调幅(Amplitude Modulation and Frequency Modulation,AM-FM)信号模型。针对以上回波信号模型,分别提出了相应的成像新算法用于提高机动目标ISAR成像的质量。本文的主要工作如下:1.详细分析了机动目标回波信号特点,并结合仿真数据分析了目标运动对回波信号的多普勒频率和图像质量的影响。同时分析了机动目标ISAR成像的特点。本文对机动目标的回波信号进行了详细的归纳和分析。2.针对LFM信号模型,文中详细分析了基于匹配傅里叶变换(Match Fourier Transform,MFT)算法的LFM信号参数估计方法。该方法具有参数估计精度高的优点。同时将此方法用于机动目标ISAR成像中,有效的解决了机动目标方位向散焦问题,提高了成像的质量。并且在一定噪声存在的情况下,算法依然有效。3.针对MFT算法计算量大的缺点,文中引入了基于二次相位函数(Quadratic Phase Function,QPF)的LFM信号参数估计方法。该方法采用一维搜索,计算量大为减少,同时该算法具有很高的参数估计精度。最后将QPF用于机动目标ISAR成像中,显著提高了成像质量。并且在一定噪声存在的情况下,算法依然有效。4.针对三次PPS信号模型,文中提出了基于扩展匹配傅里叶变换(Extension Form of Match Fourier Transform,EMFT)算法的参数估计新方法。与MFT算法相比在计算量相当的情况下保持较高的参数估计精度。同时基于一阶扰动分析原理详细推导了该方法的渐进统计特性,并通过仿真实验验证了理论推导的正确性。最后将该方法用于机动目标ISAR成像中,提高了成像的质量。5.针对四次PPS信号模型,文中提出了一种基于乘积型时间——相位导数分布(Product Time Phase Derivative Distribution,PTPDD)的多分量PPS信号参数估计新方法。该方法可有效抑制交叉项的影响,因此可以准确估计信号各个分量的参数。并且该算法估计参数时只需一维搜索,计算量小。将该方法用于复杂运动机动目标ISAR成像中,并结合距离瞬时多普勒(Range Instantanneous Doppler,RID)算法,可以得到高质量的机动目标ISAR瞬态图像。6.文中研究了任意阶次PPS信号的时频分布问题,提出了修正S分布(Modified S-Distribution,MSD)来抑制高次相位对时频聚集性的影响。MSD是S分布(S-Distribution,SD)的改进。SD在分析线性调频信号时仅在瞬时频率方向上有最佳时频聚集性。而修正SD在分析三次以上PPS信号时具有最佳的时频聚集性。当分析多分量信号时,可采用CLEAN技术逐次分离出每个信号分量。最后将该方法用于机动目标ISAR成像中,成像质量得到显著改善。7.提出了基于自适应Chirplet分解的船载(Shipborne Inverse Synthetic Aperture Radar S-ISAR)成像新方法。针对S-ISAR成像情况下目标散射点回波信号形式非常复杂的特点,本文将其刻划为多分量调频调幅信号,进而采用自适应Chirplet分解的方法对其进行分析。信号分解是一种常用的时频分析方法,它是将信号分解成一系列基函数的线性组合,然后可通过分析较简单的基本信号及它们之间的关系来达到信号分析和处理的目的。最后将该方法用于S-ISAR成像中,获得了高质量的S-ISAR图像。更多还原

【Abstract】 Synthetic aperture radar is a high resolution microwave imaging radar and has the characteristics of all-weather、day/night and long distance, widely used in the fields of national security and economy. Based on the operating mode, it can be divided into Synthetic Aperture Radar and Inverse Synthetic Aperture Radar. In the field of ISAR imaging of maneuvering targets, as to the smooth movement targets, the received signal could be represent by linear frequency modulated(LFM) signal mode, however as to the target with complex motion, the echo signal should be characterized as the model of multi-component polynomial phase signal(PPS) that is typical non-stationary signals. In the application, the quality of the radar images of targets depends on the signal processing technique, so the quality of the radar images can be improved greatly by the signal processing methods. This paper divides the echo of the maneuvering targets into four categories from the aspects of parameter estimation, time frequency analysis and signal decomposition, they are multi-component linear frequency modulated signal mode, multi-component polynomial phase signal mode, multi-component fourth-order polynomial phase signal and multi-component amplitude modulation and frequency modulation signal model respectively. For these four echo signal models, the corresponding signal processing method is presented to improve the quality of ISAR images of maneuvering targets. The main work of this paper is summarized as follows:1. Detailed analysis of the characteristics of the echo signal that from maneuvering targets and analysis of the effect of target motion on Doppler and radar images combine with simulated data. Finally analysis of the characteristics of ISAR imaging of maneuvering targets. The characteristics of the echo signal of the targets which have a maneuvering motion are summarized and analyzed in detail.2. For the linear frequency modulated(LFM) signal mode, the algorithm for estimating the parameters of LFM signal based on the Matched Fourier transform is analyzed in detail. This method has the advantage of high precision parameter estimation. Finally this method is applied to ISAR imaging of maneuvering targets, effective solve the problem of azimuth defocusing and improves the images quality greatly. Moreover, in case of the existence of a certain noise, this method is still valid.3. Due to the deficiency of MFT, a new algorithm for estimating the parameters of LFM signal based on the Quadratic Phase Function(QPF) is quoted. This algorithm can estimate the parameters of the LFM signal for each component with only one-dimensional maximization and the amount of calculation reduced greatly. QPF algorithm has the advantage of high precision parameter estimation. Finally, this method is applied to ISAR imaging of maneuvering targets and improved the images quality greatly. Moreover, in case of the existence of a certain noise, this method is still valid.4. For the cubic phase function, a new algorithm for estimating the parameters of CPF signal based on the Extension Form of Match Fourier Transform is proposed. Compared with the MFT algorithm, this method maintains a high estimation accuracy and does not increase calculation, the asymptotic statistical performance is analyzed by the first order perturbation analysis of maxima of random functions. The theoretical and simulated results demonstrate high precision of this algorithm. Finally this new algorithm is used in ISAR imaging of maneuvering ship targets and combine with the CLEAN technique to separate different signal components successfully, the radar images quality improved greatly.5. For the fourth-order polynomial phase signal, a novel method for estimating the parameters of multi-component polynomial phase signal based on the Product Time Phase Derivative Distribution(PTPDD). The PTPDD algorithm can effectively suppress the influence of cross-terms and can estimate the parameters of the PPS for each component with only one-dimensional maximization. Finally this new method is applied in the ISAR imaging of complex motion ship targets and combined with the Range-Instantaneous-Doppler(RID) technique, the high quality instantaneous ISAR images can be obtained, especially under the existence of a certain noise, this method is still valid6. The time frequency analysis of the high order polynomial phase signal are considered and the Modified S-Distribution(MSD) is presented in this dissertation. MSD algorithm is improved from SD, for linear frequency modulated(LFM) signal, the SD is only completely concentrated along the instantaneous frequency of it. Compared with SD, the MSD has better time frequency resolution in the analysis of cubic polynomial phase signal. As to the multi-component signals, the MSD can be cross-terms free based on the CLEAN technique. Finally the MSD is used in the ISAR imaging of maneuvering target, in the azimuth direction to replace the Fourier transform and the image quality is improved greatly.7. A new method of Shipborne Inverse Aperture Radar imaging for the maneuvering targets based on adaptive Chirplet decomposition is proposed. Due to complexity of the scatterer’s echoes from the target in S-ISAR imaging scenario this paper will describe it as multicomponent AM-FM signals and the signal decomposition technique based on the Chirplet basis function is adopted to deal with it. Signal decomposition is a common time-frequency analysis technique, by decomposing the signal into a linear combination of a series of basic functions, and then achieve the purpose of signal analysis and processing by analyzing the simple basic signal and their connection. Finally this method is used in the ship-borne ISAR imaging of ship target, the high quality instantaneous ISAR images can be obtained,, especially under the existence of a certain noise, this method is still valid.更多还原