铝基板表面氧化铝层分子自组装活化法镀铜

采用分子自组装活化法在铝基板表面氧化铝上实施化学镀铜,镀层有很好的剥离强度,通过SEM、EDS和离子色谱对其进行了表征。研究了硅烷化时间与基体表面硅烷修饰量的关系,浸钯时间对基体表面钯含量的影响,硅烷化时间对镀层剥离强度的影响。通过正交试验得到最佳工艺条件:硅烷化处理用3-氨基丙基三乙氧基硅烷,质量分数为0.4%,硅烷化时间12h、温度50℃、浸钯溶液活化时间30min、30℃;得到的镀铜层剥离强度为1.00。     

光电靶的设计与改进

为了测量枪、炮子弹的速度,提出一种基于光电转换的原理光电靶的设计方法。当光电靶工作时,火光或是环境光会影响光电靶产生误触发信号或是不能捕捉住弹丸过靶信号,影响了测试的准确度和灵敏度。采用新型的大面积激光光幕技术设计了光电补偿电路,将其干扰光引起的变化的光电流信号补偿掉,分析其原理,并用实验方法验证了设计的正确性。最后实弹测试其系统的有效性与实用性。结果表明,新型的补偿电路测速系统能够有效地抑制干扰光的影响。     

商业生态系统的扩展性研究

提出一个两层的商业生态系统模型,上层网络中的用户进行商业活动,构成一个生态系统;下层网络由多个查询服务器组成,构成分布式查询系统,供上层用户注册其服务代理以及查找服务.上层和下层网络随时间不断演化且相互影响.在该模型的基础上,研究下层分布式系统的可扩展性.实验表明,该系统是可扩展的,并通过比较得出影响可扩展性的因素.     

流形学习算法分析及在人脸数据库上的应用

为了更高效的处理高维数、高复杂性的非线性数据,发现其嵌入在源数据空间中的本维特征,提出了基于局部光滑逼近思想的流形学习算法,通过局部线性误差逼近最小化,实现将高维数据映射到低维空间.在FREY人脸数据库上进行降维实验,证明了该方法的可行性和有效性.     

认知中继网络中的鲁棒传输

In cognitive relay networks, the transmission of Secondary Users (SUs) suffers from the activity of Primary Users (PUs) and wireless channel fading. Therefore, how to achieve robust transmission for cognitive relay networks is a challenging task. In this paper, we propose a relaying transmission scheme which exploits robust beamforming at the physical layer and rateless codes at higher layers. We derive the optimal beamforming weight vector and analyze the performance of the proposed scheme when the channel estimation is not accurate between two SU nodes. We also study how the amount of channel information between SU and PU impacts the system performance. Simulation results validate our theoretical analysis and demonstrate that the proposed scheme can significantly enhance the system throughput.     

High-efficiency red electroluminescence from europium complex containing a neutral dipyrido(3,2-a:2′,3′-c)phenazine ligand in PLEDs

In order to obtain high-efficiency monochromatic red emission in polymer light-emitting devices, a tris(dibenzoylmethanato)(dipyrido(3,2-a:2′,3′-c)phenazine) europium [Eu(DBM)₃(DPPZ)] doped single-emissive-layer devices were fabricated using a blend of poly(9,9-dioctyl-fluorence) and 2-tert-butyl-phenyl-5-biphenyl-1,3,4-oxadiazole as a host matrix by solution process. Significantly improved electro-luminescent properties with sharp red emission at 611.5 nm were displayed in the Eu(DBM)₃(DPPZ)-doped devices at dopant concentrations from 1 to 8 wt.%. The highest luminance up to 1783 cd/m² at 2 wt.% dopant concentration, as well as the maximum external quantum efficiency of 2.5% and current efficiency of 3.8 cd/A were obtained at 1 wt.% dopant concentration.     

An efficient anonymous communication protocol for wireless sensor networks

Anonymous communication is very important for many wireless sensor networks, because it can be used to hide the identity of important nodes, such as the base station and a source node. In sensor networks, anonymous communication includes several important aspects, such as source anonymity, communication relationship anonymity, and base station anonymity. Existing sensor network anonymous schemes either cannot achieve all the anonymities or have large computation, storage, and communication overheads. In this paper, we propose an efficient anonymous communication protocol for sensor networks that can achieve all the anonymities while having small overheads on computation, storage, and communication. We compare our anonymous communication protocol with several existing schemes, and the results show that our protocol provides strong anonymity protection and has low overheads. Copyright © 2011 John Wiley & Sons, Ltd.     

部分并行软干扰消除迭代多用户检测算法研究

本文提出了一种基于部分并行干扰消除(PPIC)结构的迭代多用户检测算法,并对基于PPIC的迭代多用户检测器的迭代结构进行了改进.由于PPIC多用户检测器前一级输出的信息的准确度会影响到后一级的检测,所以为了提高基于PPIC的迭代多用户检测器的性能,对PPIC每一级的输出信息结合最大后验概率(MAP)信道译码器单独进行迭代处理,这样就提高了PPIC每一级输出的信息的准确性.该迭代结构可用于其他基于PIC型的迭代多用户检测算法.     

高频感应加热电源模数结合连续可调移相电路研究

为减小开关损耗和防止开关管损坏,通常要求高频感应加热设备工作在弱感性状态,为此必须在反馈回路进行相位的超前、滞后调节。传统的高频模拟移相电路存在移相角随输入信号频率变化时刻发生变化的缺点,而高频数字移相电路在移相精度要求较高时倍频难以实现。为此,文中提出了两种实用的高频模数结合移相电路实现移相角在任意需要角度范围内连续可调移相。经实验,在光伏电池片组件1 MHz高频感应焊接时满足移相要求,移相效果好。     

遥感图像的NSCT自适应阈值去噪方法

提出了一种基于非下采样Contourlet变换(NSCT)相结合的遥感图像自适应阈值去噪方法。通过小波估计被噪声污染遥感图像的噪声强弱,根据噪声的强弱以及NSCT的分解特点及系数所在邻域的特性,给出不同尺度不同方向的自适应阈值。仿真实验结果表明,与小波硬阈值、Contourlet硬阈值和基于非下采样Contourlet硬阈值去噪方法比较,该方法在提高了图像的峰值信噪比的同时也减少了Gibbs现象,图像视觉效果也得到明显的改善。