二维多波束蜂窝网覆盖方法研究

提出一种基于二维多波束形成的蜂窝网覆盖方法,该方法旨在通过对平面阵阵元权矢量的控制,使所形成的波束在空域上呈非均匀蜂窝网结构,达到重点地区覆盖密度高,非重点区域覆盖密度低的目的,减小资源利用率,并重点分析了波束设计思路和权值的计算方法,仿真结果表明:该覆盖方法是有效的.     

用SY87721实现城域网波分复用系统的多业务接入

为了使城域网的DWDM（密集波分复用）系统可以接入多种信号速率和格式，以满足用户的要求，首先介绍了一种波长转换板的设计方法，使用SY87721自适应提取28Mbit／s～2.7Gbit/s之间任意速率信号的时钟，其次介绍了SY87721提取时钟的工作原理、与单片机和CPLD（复杂可编程逻辑器件）的接口方式、对寄存器的配置和读出方式。最后介绍了使用参考时钟测量接入信号频率的方法，这会在系统维护时带来较大的方便。     

单液晶光阀彩色大屏幕投影显示

设计一电路,将彩色图象信息编码到黑白CRT的亮度之中,并将其写入液晶光阀,用白光读出,当编码合适时,读出的象颜色可与原图像一致。本文讨论了这一技术原理并进行了单液晶光阀彩色大屏幕投影的实验验证。     

Vibrational Mode in Two-Dimensional Dust Monolayer with Finite Size Dust Grains

Vibrational mode in a two-dimensional dust monolayer is investigated by considering the finite size of dust grains. Each dust grain is assumed to be a negative point charge and a dipole moment due to the inhomogeneous charge distribution on its surface. The dispersion relation of the vibrational mode is derived. Both the self-excited and externally excited cases are discussed. It is shown that the mode is sensitive to the direction of the dipole moment.     

Data-driven facial expression synthesis via Laplacian deformation

Realistic talking heads have important use in interactive multimedia applications. This paper presents a novel framework to synthesize realistic facial animations driven by motion capture data using Laplacian deformation. We first capture the facial expression from a performer, then decompose the motion data into two components: the rigid movement of the head and the change of the facial expression. By making use of the local-detail preserving property of the Laplacian coordinate, we clone the captured facial expression onto a neutral 3D facial model using Laplacian deformation. We choose some expression “independent points” in the facial model as the fixed points when solving the Laplacian deformation equations. Experimental results show that our approach can synthesize realistic facial expressions in real time while preserving the facial details. We compare our method with the state-of-the-art facial expression synthesis methods to verify the advantages of our method. Our approach can be applied in real-time multimedia systems.     

C/S模式PDM在网络分布式环境下集成应用研究

通过分析企业信息化网络应用过程中PDM系统所面临的网络集成问题,提出了一种基于C/S模式开发的PDM系统转化为网络分布式集成应用的解决方案,并利用中间件技术研制了实现企业应用集成的网络集成平台。基于此平台运行的各应用模块不仅可以相互间实现紧密集成,并能够基于平台资源定向功能实现应用的网络分布和W eb化访问,且对国产数据库应用提供专门支持。     

Calcium binding mode of {gamma}-carboxyglutamic acids in conantokins

Conantokin-T (con-T) and conantokin-G (con-G) are two highlyhomologous peptide toxins found in Conus venom. The former isa 21-residue peptide with four -carboxyglutamic acid (Gla) residues(at positions 3, 4, 10 and 14), while the latter is a 17-residuepeptide with five -carboxyglutamic acid residues (at positions3, 4, 7, 10 and 14). Despite the apparent similarity in numberand relative positions of the -carboxyglutamic acid residues,¹¹³Cd-NMR studies indicated a distinct metal binding behaviorfor con-G and con-T. There appears to be four binding sitesin con-G in contrast to one metal binding site in con-T. Toelucidate the mode of calcium binding by the -carboxyglutamicacid residues in these conantokins, we designed various analogouspeptides with their -carboxyglutamic acid replaced by otheramino acid residues. ¹¹³Cd-NMR experiments on conantokin analoguesreveal that the major difference in the number of metal bindingsites between con-G and con-T is due to the residue at position7. We also performed molecular simulations to calculate therelative binding free energies of several potential bindingsites. Based on our theoretical and experimental results, wepropose a `four-site' binding model for conantokin-G and a `single-site'binding model for conantokin-T.     

最优知识传递宽残差网络输电线路螺栓缺陷图像分类

目的 输电线路螺栓图像具有分辨率低和视觉信息较差的特点,针对螺栓缺陷图像分类时教师网络（大模型）参数量大、学生网络（小模型）分类精度低的问题,提出了一种最优知识传递宽残差网络输电线路螺栓缺陷图像分类方法,来弥补使用大小模型进行分类任务的局限性。方法 首先改变大模型宽度,即拓宽网络特征表达维度来增加向小模型传递的螺栓缺陷知识和简化小模型结构至3个残差块;然后为了选出传递螺栓缺陷知识性能最优的大模型,提出知识偏差的概念来可视化大模型向小模型螺栓缺陷知识传递的程度,综合分析不同宽度大模型利用知识蒸馏算法与注意力转移算法分别指导小模型训练后的分类准确率,并用知识偏差来确定最优知识传递模型;最后将最优知识传递模型利用知识蒸馏算法与注意力转移算法相结合指导小模型训练,尽可能提升小模型的螺栓缺陷分类性能。结果 在自建螺栓缺陷图像分类数据集上进行了验证,结果表明宽度为5的大模型向小模型传递螺栓缺陷知识性能最优,使小模型的螺栓缺陷分类准确率提高了5.56%,小模型与最优大模型的分类准确率只差2.17%,知识偏差为0.28,且小模型的参数量仅为大模型参数量的0.56%。结论 本文提出的最优知识传递宽残差网络输电线路螺栓缺陷图像分类方法,弥补了大小模型螺栓缺陷图像分类的局限性,实现了精度与资源消耗的平衡。     

A Cluster-Based Virtual Cooperative MIMO Transmission Scheme

基于分簇的虚拟协作MIMO传输策略研究

吕红芳

（上海电机学院 电气学院）

中文摘要：

无线传感器网络应用于环境状态监测时,最关键的问题是在无线通信时保障数据传输的可靠性、降低全网的能量消耗。本文提出了一种基于分簇的虚拟协作MIMO传输策略（CVCM）,该方案根据簇头节点到汇聚节点的距离动态配置最佳的协作节点数目和调制速率,保证每个虚拟MIMO单元和汇聚节点的通信能耗最小,从而降低整个网络的能力消耗。仿真实验结果表明,CVCM有效地降低了网络能耗,延长了网络的生命周期,提高了基站接收数据的有效性。

关键词：分簇;虚拟协作MIMO;动态配置;调制速率

     

在书写任务中的基于轨迹匹配的模仿学习

针对书写任务中运动轨迹较复杂的问题,引入基于轨迹匹配的模仿学习算法对书写轨迹进行表征和泛化,进而实现机器人书写技能的获取。机器人从示教者处获取示教数据,利用高斯混合模型( Gaussian mixture model, GMM)进行编码,学习示教行为的本质特征,通过高斯混合回归进行泛化处理,实现行为再现。实验结果表明：该方法具有良好的行为编码能力和抗干扰性,能够实现轨迹可连续的汉字书写,通过对GMM的扩展能够进行多任务学习,进而实现轨迹不可连续汉字的书写,泛化效果较好。