硅基红外焦平面阵列技术的新进展(Ⅱ)

主要介绍以硅为衬底的ＰｂＳ、ＰｂＳｅ、ＰｂＴｅ、ＰｂＳｎＳｅ、ＨｇＣｄＴｅ、ＩｎＳｂ等红外探测器焦平面阵列的新进展。     

上反式光电周视探测装置目标全方位解算方法

提出一种针对水面目标的两轴上反式光电周视探测装置目标大地全方位解算方法,该方法定义一个虚拟水平或铅直入射光平面,利用光电探测装置的航向、纵摇、横摇、方位及俯仰机构转角位置量和目标一维（水平或铅直）视频偏差量,解算水平或铅直入射光平面经上反镜反射后在探测器焦平面上像线的交点,令该交点的水平或铅直坐标等于目标视频水平或铅直偏差量,可实时解算目标以装置为原点的真北方位角.经试验证明,该方法能够有效解决光电周视探测装置自主化、小型化的技术瓶颈,使光电探测装置的功能得以提升.     

Parameter design using the Taguchi methodology

This article aims to give an overview of the Taguchi methodology and more specific of its key step: parameter design. The purpose of parameter design is to bring quality into the product/process by determining the relevant parameters and their optimal settings such that the quality characteristics of the product are optimised and have minimal sensitivity to “noise” (i.e., those disturbing factors which are difficult or too expensive to control). The (mainly statistical) tools used in the methodology will be handled and some actual Bell cases will be mentioned as practical examples.     

Efficient computation of the 2-D Green's function for 1-D periodic line sources in planar multilayered media

Periodic Green's function (PGF) of an infinite 1-D periodic array of line sources in planar multilayered media is calculated by the use of the of the 3-level discrete complex image method (DCIM) together with the surface wave pole (SWP) extraction. The Ewald's method is then applied to accelerate the evaluation of the PGF. The recipe to choose the Ewald's splitting parameter is also given. It is shown that the Ewald's transformation needs a few terms to approximate the PGF with high accuracy. The SWP contributions, which are extracted before the DCIM application, are added to this approximation. The spatial domain surface wave pole contributions are observed to be highly oscillatory and converge to zero. The numerical results are provided to demonstrate the efficiency and the accuracy of the procedure followed.     

基于容器对象的动态控件数组研究

动态控件数组是应用程序设计中经常用到的重要手段之一。合理地使用动态控件数组不但能简化程序、方便维护,还可以节约内存空间,对动态控件数组的研究有助于增加应用程序灵活性,能有效提高编程效率。鉴于此,提出了在C＋＋Builder环境下基于容器对象的动态控件数组实现方案,分别给出利用三种容器对象TList,DynamicArray和Vector实现动态控件数组的思想,对三者的性能进行比较分析,并总结了实现动态控件数组的原理、方法和技巧。     

A 640 by 480 Pixels Readout Circuit for IR Imaging

A readout circuit for a 640 × 480 pixels FPA (focal plane array) has been successfully designed, fabricated and tested. The circuit solution is based on a per pixel source-follower direct injection (SFDI) pre-amplifier. Signal multiplexing is performed in both X and Y direction. The pixel size is 25 m × 25m. The chip is optimized for a QWIP (quantum well infrared photodetector) operating at a temperature of 70 K. The circuit has been realized in a standard 0.8 m CMOS process.     

Process simulation for HgCdTe infrared focal plane array flexible manufacturing

The strategy and status of a process simulator for the flexible manufacture of HgCdTe infrared focal plane arrays is described. It has capabilities to simulate Hg vacancy and interstitial effects and cation impurity diffusion, for various boundary conditions in one dimension. Numerical complexity of these problems stems from the necessity of solving diffusion equations for each defect that are coupled to each other via nonlinear interaction terms. The simulator has already led to the prediction of heretofore unexplained experimental data. Current extensions of the one-dimensional simulator planned over the next few years include the addition of Te antisites, antisite-Hg vacancy pairs, and In-Hg vacancy pairs, ion implantation, and various energetic processes (such as ion milling). The sequential effect of various processes will be possible with the input to the simulator looking much like a process run sheet.     

硅树脂调制的可调光子晶体

文章提出一种利用硅树脂调节光子晶体光子带隙的方法,光子晶体波导是通过往二维正方形光子晶体的介质柱之间填充硅树脂得到的,利用温度场改变硅树脂的折射率.数值模拟结果表明:通过温度场可对这种光子晶体的禁带结构进行调节.这种可调光子晶体可应用于制作新颖的偏光片和光开关.     

Design of a Temporal Learning Chip for Signal Generation and Classification

The design and analog VLSI implementation of a recurrent neural network with integrated temporal learning is presented. The learning algorithm is forward in time, and is implemented strictly as instantaneous, local weight updates. PSpice simulations of networks with 4 to 6 neurons demonstrate robust learning of trajectory generation and classification tasks. A scalable 2-D VLSI architecture is described and a prototupe 4-neuron recurrent neural network with learning has subsequently been fabricated in MOSIS TinyChip 2 micron technology. Experimental results of the chip validate the learning performance with convergence in the millisecond range. Specific experimental results of learning circular and figure-8 dynamic trajectories are included.