A new FDTD algorithm based on alternating-direction implicit method

In this paper, a new finite-difference time-domain (FDTD) algorithm is proposed in order to eliminate the Courant-Friedrich-Levy (CFL) condition restraint. The new algorithm is based on an alternating-direction implicit method. It is shown that the new algorithm is quite stable both analytically and numerically even when the CFL condition is not satisfied. Therefore, if the minimum cell size in the computational domain is required to be much smaller than the wavelength, this new algorithm is more efficient than conventional FDTD schemes in terms of computer resources such as central-processing-unit time. Numerical formulations are presented and simulation results are compared to those using the conventional FDTD method     

The 3-D multidomain pseudospectral time-domain method for wideband simulation

A three-dimensional (3-D) multidomain pseudospectral time-domain (PSTD) method with a well-posed PML is developed as an accurate and efficient solver for Maxwell's equations in conductive and inhomogeneous media. The curved object is accurately treated by curvilinear coordinate transformation. Spatial derivatives are obtained by the Chebyshev collocation method to achieve a high-order accuracy. Numerical results show an excellent agreement with solutions obtained by the FDTD method under fine sampling.     

MPSTD算法中子域分界面匹配条件的比较研究

针对多区域伪谱时域算法(MPSTD)中子域分界面两侧为不同介质的情形,提出了特征变量-物理边界(CV-PB)子域分界面匹配条件,然后在总结现有子域分界面匹配条件的基础上,分别比较了子域分界面两侧为相同和不同介质时各种子域分界面匹配条件的稳定性.数值仿真表明,在子域分界面两侧为相同介质材料时采用特征变量法匹配条件,而在子域分界面两侧为不同介质材料时采用CV-PB匹配条件,可使得MPSTD算法稳定性最好.     

The 3-D multidomain pseudospectral time-domain algorithm for inhomogeneous conductive media

A three-dimensional (3-D) multidomain pseudospectral time-domain (PSTD) method with a strongly well-posed perfectly matched layer (PML) is developed as an accurate and flexible tool for the simulation of electromagnetic wave propagation and scattering in inhomogeneous and conductive media. This approach allows for an accurate treatment of curved geometries by multidomain decomposition and curvilinear coordinate transformation. Numerical experiments show the results agree excellently with analytical solutions and results of other well-known algorithms, and demonstrate a remarkable improvement in accuracy and efficiency over the FDTD method. The 3-D multidomain PSTD algorithm is then applied to calculate radar cross sections (RCS).     

A three-dimensional horizontally wide-angle noniterative beam-propagation method based on the alternating-direction implicit scheme

A three-dimensional horizontally wide-angle beam-propagation method is proposed on the basis of the alternating-direction implicit scheme, in which the Pade/spl acute/ approximant is applied only to the horizontal direction. The present formulation reduces the splitting error to the first order without an iteration procedure. The effectiveness is demonstrated through the wide-angle propagating beam analysis of a tilted optical waveguide.     

The unconditionally stable pseudospectral time-domain (PSTD) method

This letter presents a new time-domain method for Maxwell's equations, in which the unconditionally stable techniques, the alternating direction implicit (ADI) and the split-step (SS) schemes, are developed for the pseudospectral time-domain (PSTD) algorithm to maintain stability while achieving higher accuracy and efficiency over the FDTD method. The multidomain strategy is employed to allow for a flexible treatment of internal inhomogeneities. Numerical results demonstrate the unconditional stability and the second-order accuracy for both ADI- and SS-PSTD algorithms.     

A pseudospectral method for time-domain computation ofelectromagnetic scattering by bodies of revolution

We present a multidomain pseudospectral method for the accurate and efficient time-domain computation of scattering by body-of-revolution (BOR) perfectly electrically conducting objects. In the BOR formulation of the Maxwell equations, the azimuthal dependence of the fields is accounted for analytically through a Fourier series. The numerical scheme in the (r,z) plane is developed in general curvilinear coordinates and the method of characteristics is applied for patching field values in the individual subdomains to obtain the global solution. A modified matched-layer method is used for terminating the computational domain and special attention is given to proper treatment of the coordinate singularity in the scattered field formulation and correct time-domain boundary conditions along edges. Numerical results for monochromatic plane wave scattering by smooth and nonsmooth axis-symmetric objects, including spheres, cone-spheres, and finite cylinders, is compared with results from the literature, illustrating the accuracy and computational efficiency associated with the use of properly constructed spectral methods. To emphasize the versatility of the presented framework, we compute plane wave scattering by a missile and find satisfactory agreement with method-of-moment (MoM) computations     

基于空谱分解的时间反演时域成像

提出了基于空谱分解的时间反演时域成像新方法.通过天线阵列的一次测量,获取散射场数据、建立空谱多态响应矩阵.奇异值分解矩阵求取与具体目标相对应的左奇异矢量,与超宽带信号结合合成相干的时间反演后向传输信号,实现了目标的选择性聚焦.该方法能实现置于均匀媒质或随机媒质中的多个运动目标的聚焦成像,具有较好的实时性及较强的成像能力.     

计算电大尺寸建筑物内电波场强的PSTD方法

本文采用一种新的时域数值方法－伪谱时域（ＰＳＴＤ）法来计算电大尺寸建筑物内电波场强。提出了由初始条件技术和一维ＰＳＴＤ方程对入射平面波脉冲进行模拟,并利用纯散射场法和线性插值把平面波引入求解问题空间,有效地解决了ＰＳＴＤ方法中入射波设置问题。数值结果表明这种新方法用于模拟电大尺寸建筑物内电波场强的精度和有效性。     

基于异构隐式存储的多模式匹配算法

提出了紧缩存储型Aho-Corasick算法变体,以异构的按需隐式存储取代同构的例行显式存储,从横向扇出压缩与纵向路径压缩2个方向入手,围绕着压缩稀疏事件表展开,当字符集大小σ=256时可将存储量缩减为原来的0.69%左右,而σ＝64K时则达0.004%,即空间复杂度降为原来的(1bσ)/σ左右.依据扇出疏密程度的不同,分类采用了4种有针对性的快速事件定位方法,加之优化的失败迁移,使得存储量的大幅缩减不以速度的明显损失为代价,实验也证实了这一点.适用于需承载大型模式集和较长模式串而对时延和抖动都比较敏感的场合(如在线数据流过滤),在宽字符(如UNICODE型亚洲字符)匹配方面拥有显著优势.     

基于隐式数据和Apriori的协同过滤推荐算法

针对传统协同过滤推荐算法对目标客户进行个性化推荐时,因用户评价数据和物品属性等显式数据稀疏,造成推荐商品的准确率和质量相对较差的问题,本文基于隐式数据和Apriori算法对协同过滤推荐算法做出改进.首先,算法基于隐式数据中用户对商品的行为和用户对商品的评价,建立用户对商品的评分偏好模型,用以构建原始评分数据;其次,利用...     

Alternating-direction implicit (ADI) formulation of the finite-difference time-domain (FDTD) method: algorithm and material dispersion implementation

The alternating-direction implicit finite-difference time-domain (ADI-FDTD) technique is an unconditionally stable time-domain numerical scheme, allowing the /spl Delta/t time step to be increased beyond the Courant-Friedrichs-Lewy limit. Execution time of a simulation is inversely proportional to /spl Delta/t, and as such, increasing /spl Delta/t results in a decrease of execution time. The ADI-FDTD technique greatly increases the utility of the FDTD technique for electromagnetic compatibility problems. Once the basics of the ADI-FDTD technique are presented and the differences of the relative accuracy of ADI-FDTD and standard FDTD are discussed, the problems that benefit greatly from ADI-FDTD are described. A discussion is given on the true time savings of applying the ADI-FDTD technique. The feasibility of using higher order spatial and temporal techniques with ADI-FDTD is presented. The incorporation of frequency dependent material properties (material dispersion) into ADI-FDTD is also presented. The material dispersion scheme is implemented into a one-dimensional and three-dimensional problem space. The scheme is shown to be both accurate and unconditionally stable.     

Efficient implicit FDTD algorithm based on locally one-dimensional scheme

The locally one-dimensional (LOD) scheme is introduced into the implicit FDTD method. The LOD-FDTD provides simple implementation of the algorithm and reduced computational time. For facet reflectivity analysis of an optical waveguide, the computational time is reduced to 80 and 50% of those of the ADI-FDTD and the conventional FDTD, respectively.     

基于遗传算法的高维子空间聚类算法设计

针对高维空间数据的特点,为了降低“维数灾难效应”对聚类结果的影响,提出并实现了一种新的基于遗传算法的子空间聚类算法,通过特征选择方法并结合遗传算法的全局搜索能力对所有的特征子空间进行搜索:采用实数制编码方式对解空间进行编码,并设计一种基于距离和信息熵的适应度评估函数来对聚类结果和子空间所包含的特征维进行评估.最后,通过人工数据与真实数据等几组实验验证了算法的高效性和鲁棒性.实验结果表明,本文提出的新算法能够有效地进行高维数据聚类,降低“维数灾效应”的影响.     

Steady-state response by finite-difference time-domain method and lanczos algorithm

A hybrid method combining the finite-difference time-domain (FDTD) method and Lanczos algorithm is proposed for efficiently obtaining the steady-state response for closed systems. With the Lanczos algorithm, modes of the FDTD operator can be easily extracted to reconstruct the steady-state response at any time by an analytic formula. With the aid of the time-reversal technique, existing FDTD codes are preserved. Numerical results show that with only very few Lanczos iterations, results of good agreement can be achieved between the proposed method and brute-force FDTD.     

Computer-aided design of two dimensional electric-type hyperthermiaapplicators using the finite-difference time-domain method

A hyperthermia applicator design tool consisting of a finite-difference time-domain (FDTD) technique in combination with a graphical display of electric fields and normalized linear temperature rise is described. This technique calculates, rather than assumes, antenna current distributions; it includes mutual interactions between the body and the applicator, and it calculates driving-point impedance and power delivered to the applicator. Results show that the fundamental limitation of 2-D electric-type applicators is overheating of the fat by normal components of the electric field, which exist because of near fields and capacitive coupling with the muscle. Two factors which contribute to the capacitance are the muscle conductivity and the small antenna size in air. Two examples of applicators designed to avoid fat overheating are described: a 27-MHz segmented dipole for heating large tumors to 7 cm depth, and a 100-MHz dipole for small tumors to 5 cm depth. The first uses a water bolus, and the second uses a water bolus with low-permittivity strips to reduce normal fields at the antenna ends. The results of this study describe fundamental limitations of electric field applicators, and illustrate the use of a powerful applicator design tool that allows rapid evaluation of a wide range of ideas for applicators which would require months and years to test experimentally.     

一种基于时域混合的DWT-SVD盲数字水印算法

提出了一种基于时域混合的盲数字水印新算法。该算法一改传统水印的嵌入过程,首先对载体图像进行分块,利用最佳信号与载体图像进行时域混合以改变图像信息的分布,之后通过离散小波变换结合奇异值分解完成水印嵌入,再进行时域混合恢复得到嵌入水印的图像。进行各种攻击测试,并与传统DWT-SVD数字水印算法进行比较。实验结果表明,时域混合算法具有很好的隐蔽性和鲁棒性。     

一种基于时域滤波的BOC无模糊捕获算法设计

针对二进制偏移载波(BOC)信号进行相关运算时存在多峰的特殊性,设计了一种基于时域滤波的BOC无模糊捕获算法。通过频域分析得到时域滤波器的频率响应,再反推出时域滤波器的系数,构建有限脉冲响应(FIR)滤波器。对不同算法进行仿真,结果证明该时域滤波器能很好地消除BOC信号自相关副峰。降低滤波器阶数,并设计可实现的基于时域滤波和多段匹配滤波器(MMF)－傅里叶变化(FFT)的BOC无模糊捕获算法,仿真分析表明该算法能很好地消除副峰,正确捕获信号。     

基于隐式Gabor滤波与PCA的人脸识别算法

通过卷积计算Gabor特征然后再进行主成分分析(PCA)是经典的人脸识别方法之一,能取得较高的识别率。但由于需要计算多角度、多尺度下的Gabor特征,所以该方法的效率很低。为了提高效率,文中提出了一种利用Gabor特征但并不直接计算Gabor特征的主成分分析的方法。所提方法将Gabor特征的信息隐含在一个离线计算好的矩阵中,并将该矩阵包含在主成分特征提取中。在FERET和YALE人脸数据库上的实验表明,所提方法的识别速度提高了5倍左右,识别率也略有提升。