摘要
摘要霜儿九岁
雷达信号处理中，雷达目标参数估计是雷达目标探测的重要任务之一。传统的多普勒频率估计，雷达分辨力是由相干积累时间决定的，较长的相干积累时间可以获得较高的分辨力，但受雷达系统工作模式、目标自身特性以及电磁传播环境等限制，相干积累时间有限。因此，在有限相干积累时间下快速获取目标信号多普勒频率的高精度估计时，经典方法遇到了瓶颈。稀疏表示理论的出现为解决这一问题带来了新的曙光。
本文主要研究了稀疏表示理论在雷达目标参数估计方面的应用, 由于基本原理相同，本文主要以多普勒频率估计为例进行了研究。首先，介绍了信号稀疏表示的基本理论；其次，建立了信号多普勒频率估计的稀疏模型，并通过经典的贪婪类迭代重构算法进行频率估计，获得了较高的分辨力；最后，在前述研究的基础上，针对天波超视距雷达回波的多径效应，提出了一种基于稀疏表示的海杂波一阶峰匹配算法，改善了天波超视距雷达在强杂波背景下对海面舰船目标的检测性能。
本文利用雷达信号通常在多普勒频域具有稀疏性的特点，在短相干积累时间下基于系统的先验信息对观测目标场景进行稀疏建模，在系统建模的基础上，对目标回波信号模型进行了推导。针对稀疏度先验信息已知的回波模型，本文利用了正交匹配追踪算法以及减少了迭代次数的正则化正交匹配追踪算法对目
标的多普勒频率进行估计，通过仿真实验可以得出，其精度要比传统的傅里叶分析法方法高。对稀疏度先验信息未知的回波模型，本文利用了稀疏度自适应的匹配追踪算法对目标的多普勒频率进行估计，该算法通过自适应的调整迭代步长来逐渐对信号的稀疏度进行估计。在短相干积累条件下，天波超视距雷达舰船目标检测中海杂波的遮蔽问题，工程上的解决方法主要是海杂波循环对消法和高分辨谱估计法。本文分别使用Root法，FFT相位分析法和CAPES方法进行海杂波循环对消，实现了海杂波的抑制。当雷达回波信号经过电离层发生多径效应时，一个距离单元内的多普勒谱中会出现多对Bragg峰，传统的海杂波循环对消和高分辨谱估计法均无法将目标和多个海杂波一阶谱峰区分开来。针对上述问题，本文开展了一种基于稀疏表示的海杂波一阶峰匹配算法的研究，该方法利用同一传播模式下的正负一阶峰多普勒频率差恒定的特性对海杂波一阶谱峰配对从而区分舰船目标和海杂波一阶谱峰，实现舰船目标的检测，通过实验仿真分别验证了算法的有效性。
关键词：参数估计，稀疏表示，海杂波抑制，舰船检测，多径效应
ABSTRACT
恋爱小技巧
ABSTRACT
In radar signal processing, radar target parameter estimation is one of the important task of radar target detection. For Traditional Doppler frequency estimation, radar resolution is determined by the
coherent integration time, longer coherent accumulation time can get a higher resolution, but is limited by the radar system operating mode, the target's own characteristics and electromagnetic propagation environment, etc. Coherent accumulation time is limited. Therefore, when the high-precision estimation of the Doppler frequency of the target signal is quickly obtained under the limited coherent accumulation time, the classical method encounters a bottleneck. The appearance of spar reprentation theory brings a new dawn to solve this problem.
This paper mainly studies the application of spar reprentation theory in radar target parameter estimation. Due to the same principle, this paper mainly studies Doppler frequency estimation as an example.Firstly, the basic theory of signal spar reprentation is introduced. Secondly, a spar model of signal Doppler frequency estimation is established, and a classical greedy iterative reconstruction algorithm is ud to estimate the frequency, then a higher resolution is obtained; Bad on the foregoing rearc,aiming at the multi-path effect of Skywave Over-The-Horizon Radar, a a clutter first-order peak matching algorithm bad on spar reprentation was propod, which improved the detection performance of a-ship target by Skywave Over-The-Horizon Radar .
In this paper, radar signals are usually spar in the Doppler frequency domain. Bad on the prior information of the system, spar modeling of the obrvation target scene is performed under the s
hort coherent accumulation time. On the basis of the system modeling, the target echo is obtained. The signal model was deduced. For the known echo model with spar prior information, this paper us the orthogonal matching pursuit algorithm and regularized orthogonal matching pursuit algorithm with reduced number of iterations to estimate the Doppler frequency of the target, and it can be obtained that the accuracy is higher than the traditional Fourier analysis method through simulation experiments. For the unknown echo model with spar prior information, the spar adaptive matching pursuit algorithm is ud to estimate the Doppler frequency of the target. Gradually ,the algorithm adjusts the sparsity of the signal by adaptively adjusting the
西安电子科技大学硕士学位论文
iteration step size to make an estimate. Under the condition of short coherent accumulation, the shielding problem of a clutter in the Skywave Over-The-Horizon Radar target detection is mainly solved by the a clutter cycle cancellation and high-resolution spectrum estimation method. In this paper, the a clutter cycle cancellation is realized by using Root method, FFT pha analysis method and CAPES method respectively, and the a clutter suppression is realized. When the radar echo sign al undergoes multipath effect in the ionosphere, multiple pairs of Bragg peaks appear in the Doppler spectrum in a distance unit. The conventional a clutter cycle cancellation an怎样训练口才
d high-resolution spectral estimation methods cannot parate the target and Multiple a clutter’s first-order spectral peaks. In order to solve this problems, a spar reprentation bad on first-order peak of a clutter matching algorithm is studied. The method us the characteristics of the constant difference of the positive and negative first-order peak Doppler frequency in the same propagation mode for a clutter. The spectral peak pairing distinguishes the ship target and the first-order spectral peaks of the a clutter to achieve the detection of ship targets. The effectiveness of the algorithm is verified by experimental simulations.
Keywords: Parameter Estimation, Spar Reprentation, Sea Clutter Suppression, Ship Detection, Multipath Effect
插图索引
图2.1过完备稀疏表示示意图 (11)
图2.2
l范数的曲线示意图 (12)
性吧有声小说p
图3.1相邻M个探测周期的接收信号 (20)
图3.2MP算法执行流程图 (23)
图3.3OMP算法的目标多普勒估计 (25)
图3.4频率估计相对误差随M的变化曲线图 (25)
图3.5频率估计相对误差随SNR的变化曲线图 (26)
图3.6频率估计成功概率随M的变化曲线图 (27)
人体作画图3.7ROMP算法的目标多普勒估计 (28)
图3.8频率估计相对误差随M的变化曲线图 (29)
图3.9频率估计相对误差随SNR的变化曲线图 (29)
图3.10SAMP算法的目标多普勒估计 (31)
图3.11频率估计相对误差随M的变化曲线图 (32)
图3.12频率估计相对误差随SNR的变化曲线图 (32)
泣不成声
图4.1海洋回波的多普勒谱 (35)
图4.2模拟海杂波的功率谱 (41)
图4.3加入模拟目标的海面回波谱 (42)
图4.4仿真海杂波数据的循环对消结果 (43)
图4.5多径现象的产生原理图 (44)
图4.6模拟多径效应的功率谱 (44)
图4.7检测流程图 (47)网段
图4.8仿真数据的距离-多普勒图 (48)
图4.9回波信号的稀疏估计频谱图 (48)
图4.10仿真数据Bragg峰频差分布图 (49)
老婆的生日祝福语
图4.11仿真数据目标待检测结果 (50)
表格索引
的各种算法对比 (13)
表2.1 求解f Dc
