拉盖尔-高斯光束通过有硬边光阑光学系统的传输

采用将硬边光阑展为有限个复高斯函数之和的方法,得到了拉盖尔-高斯光束通过有硬边光阑近轴ABCD光学系统中传输的近似解析公式,作了数值计算以说明传输公式的应用,并对使用该式的误差作了讨论。     

一类非线性方程解的存在和唯一性及在图像中的应用

本文我们给出一个修正的非线性扩散方程模型，与Cotte Lions和Morel的模型相比该模型有许多实质上的优点。主要的想法是把原来去噪声部分：卷积Gauss过程替代为解一个有界区域上的线性抛物方程问题，因此避开了对初始数值如何全平面延拓的问题。我们从数学上的证明该问题解的存在性和适定性，同时给出对矩形域情况的解的级数形式。最后我们给基于本模型的数值计算差分模型，并且给出几个具体图像在该模型下处理结果。     

DC-DC变换器突加负载时的动态性能研究

提出了一种能够有效改善DC-DC变换器动态性能的方案。该方案是在全桥电路基础上并联一单端正激变换电路，使变换器负载从10%轻载突加到100%满载时，其暂态过程小于1μs输出电压达到设计要求。文中对具体电路进行了详细分析，并通过仿真进行了验证。     

一种适于自适应光学的旋转软刀口波面传感器的研究

利用傅里叶光学理论,证明了一种新型波面传感器——旋转软刀口波面传感器的概念.在经典刀口法的刀口区,以渐变透过率代替突变透过率函数,在光瞳象面上可获得反映光瞳面位置待测波面位相梯度及位相梯度方向的调制信号.文中给出了实验结果和讨论.     

基于遗传算法的图像边缘信息链表匹配技术

提出了一种基于遗传算法的图像边缘信息链表匹配方法。首先,采用CANNY算子的方法提取图像边缘;然后,将图像边缘信息链表化;最后,基于链表化结果利用遗传算法进行链表匹配。该方法能够减少匹配过程中的计算量,且能实现高效匹配。通过仿真实验,验证了该方法的性能。     

The Content Driven Mobile Internet

The development of future mobile networks will be driven, in large part, by content and web based services. In this paper, we examine several performance, scalability and architectural challenges faced by future mobile web applications and how advanced mobile content delivery techniques can address these challenges. We review existing content delivery using a taxonomy that consists of three categories: network scaling, end system acceleration, and content and protocol optimization. While wireline content delivery focuses on network and server scalability, mobile content delivery will likely benefit most from optimizing radio link usage. We also present our ongoing work in this area, which extends the functionality of edge caching to the terminal, uses user interest correlation information to maintain low terminal power consumption and adds a new dimension to radio resource management. This revised version was published online in August 2006 with corrections to the Cover Date.     

Modification to the Fringe Component of Diffraction Coefficient in TD-EEC Method

A modified formulation for fringe component of diffraction coefficient is implemented to TD-EEC method. An example of diffraction by perfectly conducting plate is used to illustrate our scheme. Comparing with the FDTD results we observe that the improved expression for fringe component is more accurate than that of Michaeli's formulation. This high frequency time domain technique is available for treating the bistatic scattering problems for millimeter waves.     

Low-resistance polysilicon process in contact application for high density devices

For the lowest resistance, it is required to have the epitaxial silicon contact between the silicon plug and the substrate and good step coverage at the high aspect-ratio contact holes, simultaneously. In this work, a double polysilicon (DPS) deposition technique was proposed for the requirements. The first, thin silicon layer is deposited in a single-wafer process chamber with an in-situ H₂-RTP (rapid thermal process) treatment for the epitaxial contact, and the second silicon layer is formed in a batch-type furnace for good step coverage. From chain resistance, Kelvin R_c, and current-voltage (I–V) measurement, the DPS process meets both low resistance and good uniformity, so that it suggests a breakthrough in the small-sized, semiconductor device application.     

图像边缘检测在关节镜图像处理中的应用

主要论述了采用图像边缘检测和图像跟踪算法对采集的关节镜图像进行图像处理和分析的过程。在图像处理的基础上，系统自动测量出指定目标的形状参数，实现了系统的测量要求。本研究为实现关节镜手术过程图像的保存和手术的准确性提供了可行的方法。     

Strongly confined quantum wire states in strained T-shaped GaAs/InAlAs structures

The confinement energy of T-shaped quantum wires (QWRs), which were fabricated by the cleaved edge overgrowth technique in a way that the QWRs form at the intersection of In_0.2Al_0.8As stressor layers and the overgrown (1 1 0) GaAs quantum well (QW), is examined using micro-photoluminescence spectroscopy. Photoluminescence (PL) signals from individual QWRs can be spatially resolved, since the strained films are separated by 1 μm wide Al_0.3Ga_0.7As layers. We find that due to the tensile strain being transmitted to the QW, the confinement energy of the QWRs rises systematically up to 40 meV with increasing thickness of the stressor layers. By reducing the excitation power to 0.1 μW the QWR PL emission occurs 48 meV redshifted with respect to the QW. All QWR peaks exhibit smooth lineshapes, indicating the absence of pronounced exciton localization.