共查询到20条相似文献,搜索用时 125 毫秒
1.
2.
针对复杂海杂波环境下极化SAR图像弱小目标检测难题,最近提出的雷达极化对角加载滤波器融合了极化白化滤波器(PWF)和极化检测优化滤波器(PDOF)的优势,通过选择合适的线性加权系数,能够突破非高斯杂波背景下的目标检测性能限制。但是如何获取该线性组合的最优加权权重是该方法的难点。首先重新构造了基于线性组合的雷达极化对角加载检测器,在此基础上从曲线下面积(AUC)的检测性能评估角度给出了线性组合的最优化数学模型,并提出了基于该准则的最优权重算法。为加快求解速度,分别提出了基于Fisher准则线性判别分析(LDA)、基于口袋感知机学习算法(PPLA)和以LDA解为初值AUC求解算法等三种简化求解方法。最后通过仿真和实测实验验证了以上方法的有效性和鲁棒性,结果表明以LDA解为初值的AUC求解算法综合性能最佳。 相似文献
3.
4.
在天气雷达信号处理阶段,地物杂波的存在是影响数据质量的最大问题,因此,识别并剔除地物回波是天气雷达信号处理系统的一个重要内容。首先简要介绍了IIR椭圆地物杂波滤波器的原理;其次介绍了结合杂波识别(CMD)算法的自适应高斯频域滤波器(GMAP)的滤波算法和双高斯滤波器(BGMAP)算法,分析了不同回波情况下GMAP滤波器和BGMAP滤波器的滤波效果和运行时间比较;最后根据实际雷达回波信号,分析了IIR椭圆地物杂波滤波器与GMAP滤波器的杂波抑制性能,对两种滤波器结果进行了分析和比较。结果表明,BGMAP滤波器的杂波抑制性优于GMAP滤波器和IIR椭圆滤波器,然而处理时间较长,适合对回波数据的分析,实时处理时使用GMAP滤波器能满足实时处理的要求。 相似文献
5.
6.
7.
极化自适应滤波算法的新实现 总被引:2,自引:0,他引:2
当目标信号与杂波干扰信号在时域、频域和空域的状态特征难以区分时,若两者在极化域可分,则可利用极化信息进行雷达目标检测。由于在强杂波的情况下信干比比较小,而且干扰的极化状态通常是未知的,或随时间或空间是不断变化的,传统的极化滤波算法并不能满足要求。文中针对强杂波环境下,提出了一种极化域新的自适应滤波算法,该方法基于线性约束最小方差(LCMV)准则,采用变极化接收技术,实现对信号的最佳接收,为实现实时处理和自适应快速变化的干扰,推导了一种递推极化滤波算法,为实现实时跟踪提供了一种可能,仿真结果也证明了该方法的有效性。 相似文献
8.
最优MTI滤波器的权值与目标的多普勒频率有关,传统的MTI雷达需要利用一组能覆盖目标多普勒频率范围的并行滤波器组来实现最佳滤波。本文提出了一种算法,通过子空间分解法直接求取滤波器的权系数,由于无需多普勒滤波器组,系统的复杂度得到了降低。为了适应杂波的变化,本文还推导了该算法基于幂迭代的自适应过程。仿真结果表明:在不同的杂波环境下,该算法均有良好的MTI改善因子。 相似文献
9.
10.
11.
采用固相合成工艺,制备了(Bi0.5Na0.5)(1–x)CaxTiO3(BNCT)陶瓷(x=0.01~0.14)。研究了Ca含量对BNCT陶瓷介电性能的影响。结果表明,所有BNCT陶瓷样品中都存在第二相TiO2;随着Ca含量的增加,BNCT陶瓷的介电峰向低温方向移动,介温曲线越来越平坦,剩余极化急剧减小,介质损耗逐渐减小。往x=0.12的BNCT陶瓷里掺入质量分数为1.5%的MnCO3,所得陶瓷样品的室温tanδ小于1%、绝缘电阻率达到1011?.cm,并满足–55~+250℃下,?C/C25℃≤±15%的高温MLCC的要求。 相似文献
12.
极化定标是多极化合成孔径雷达(PolSAR)系统获取真实目标极化散射矩阵的必要步骤。文中针对极化定标过程中存在的极化隔离度距离向空变特性,采用基于无源角反射器的极化定标算法作为仿真工具,对PolSAR系统极化隔离度绝对值及其空变范围对图像中极化隔离度的影响规律进行了分析。分析结果表明:当用于极化修正的极化误差矩阵与PolSAR系统自身极化隔离度不匹配时,极化修正将会使得修正后的图像极化隔离度变差。因此,在进行极化定标时,应当尽可能在定标场沿距离向布设一定间距的多组定标器,从而获取PolSAR系统在整个距离向成像带宽内的极化隔离度拟合曲线,在进行极化修正时对图像中距离向不同的像素点采用曲线中对应位置的极化误差矩阵进行修正,进而得到尽可能接近目标真实极化散射矩阵的估计值。 相似文献
13.
Qi Haiming Yu Weidong Dong Qian Zhang Huansheng 《电子科学学刊(英文版)》2007,24(3):321-325
The estimation of clutter polarization is significant in virtual polarization filtering. This paper proposes variable step size Least Mean Square error (LMS) algorithm to estimate clutter polarization. The adaptive recursive function is set up based on double polarization receiving signal model. By choosing desired response signal properly, the received data are estimated. The result of estimation is used for Null Phase-Shift instantaneous Polarization Filter (NPSPF) design and target's amplitude and phase compensation. Algorithm's tracking performance is discussed in detail. The simulation results are consistent with the above analysis, and the clutter suppression is effective. 相似文献
14.
《Antennas and Propagation, IEEE Transactions on》2007,55(9):2550-2558
Finite element time domain (FETD) codes using a Floquet modal absorbing boundary condition to model scattering from periodic structures require the computation of time consuming convolution integrals. In this paper, we propose, for the first time, to reduce this computational burden using recursive convolution. Recursive convolution is based on the ability to accurately approximate functions, over the entire computation time, using a summation of exponential functions. A novel approach, based on the exponential curve fitting facility of the commercially available software MATLAB, is employed. The time efficiency of the developed FETD code based on recursive convolution is demonstrated by comparing its computational speed with that of an FETD code employing standard convolution when modelling plane wave scattering from two-dimensional singly-periodic structures. 相似文献
15.
Describes how to evaluate exactly in a recursive manner the Jones transfer matrix of a fiber for any order of polarization mode dispersion (PMD) and derives for it a second-order accurate analytical approximation. The second-order approximation is then compared against the exact result as well as with other models found in the literature 相似文献
16.
保偏光纤偏振特性测试系统的光路研究 总被引:1,自引:1,他引:0
光纤陀螺是干涉型传感器,光的偏振特性影响该传感器的精度,因此需要对光纤的偏振态进行精确的测试。目前国外有些公司出售检偏系统,但所给的指标并不高,而且价格也非常昂贵,国内使用的测试系统一般都为自制.精度普遍不高,实际测量的消光比小于系统的理论消光比。有偏振特性系统的设计和测试结果,却没有系统误差原因的详细分析。针对这一问题,在给出了高精度保偏光纤偏振测试系统的光路结构基础上.对系统中影响消光比的光路误差进行了分析和计算。通过理论分析和计算机模拟计算,得到本系统的消光比在给定的误差范围内可以达到55dB,能够满足高精度的测量。同时,在计算中得到了e光和。光的透射率,计算结果中棱镜e光的透射率比以往的结果精确2.34%. 相似文献
17.
18.
Otsu法是一个常用的阈值分割方法.为了利用图像的区域信息,本文在二维Otsu法的基础上提出了曲线阈值型Otsu法,传统的二维Otsu法可以看成是该方法的一个特例.实验结果表明,对于含噪图像,它能够获得优于传统二维Otsu法的分割效果.为了减少计算量,提高分割速度,给出了一种递归算法和一种小波变换与递归算法相结合的快速算法.该递推算法只需遍历二维直方图的主对角线和一条次主对角线,与传统Otsu法的递推算法相比,搜索空间由L×L个点减少到2L-1个点. 相似文献
19.
Polarization properties of the flint glass fiber 总被引:2,自引:0,他引:2
To realize stable characteristics of a current sensor using the Faraday effect in an optical fiber, it is necessary to prevent the influence of birefringent effect on light passing the fiber, which is caused by photo-elastic effect and/or lack of axial symmetry of its structure. The flint glass fiber manufactured from flint glass with very small photo-elastic constant, which has circular core and satisfies the single ode condition,has been developed to solve this problem. This paper reports results of a study to clarify polarization properties of the fiber and to consider a method for highly stabilizing the polarization. From experiments to investigate the relationship between state of polarization of light inserted to the fiber and that emitted from another end, it was confirmed that the birefringence due to the photo-elastic effect and the lack of symmetry are negligibly small. It also became known that the azimuth angle of polarization of the emitted light depends on the shape of the curve of the fiber. Namely, a rotation of the polarization plane is equal to the amount of line integral of the torsional rate along the curve. A countermeasure is necessary for this phenomenon, because characteristics of the current sensor may be influenced if the flint glass fiber coil is deformed by environmental factors. From a consideration of the mechanism of the phenomenon, it is inferred that the problem can be solved by reciprocating light in the fiber to stabilize the polarization. The effect of this method is confirmed experimentally 相似文献
20.
递归式MTI滤波器从理论上讲几乎可以合成任意形状的频率响应函数,因而具有很广的应用范围。为了能够灵活设计出满足系统要求的递归式MTI滤波器,提出了一种设计方法。该方法以遗传算法(GA)的基本思想为基础,利用滤波器的改善因子、通带宽度和通带波纹来构成适应函数,并用GA来加快参量搜索速度和获取全局近似最优解。实验仿真设计了两种典型的递归式MTI滤波器,结果证明该方法是一种有效的设计方法,它可以灵活而快速地设计出所需要的滤波器。 相似文献