一种适于HTTP数据还原的QS改进算法

为了更快速、准确地对HTTP应用数据进行还原,文中研究了改进的单模式匹配算法。对BM算法、BMH算法和QS算法进行了分析,并重点研究了QS算法的改进思路,最后提出了一种适用于HTTP应用数据还原的CIQS算法。CIQS算法考虑了HTTP模式串的字符特点,改进了模式串的字符比较顺序,并对坏字符跳跃思想进行了改进,增大了跳跃距离。实验结果表明,CIQS算法有效减少了匹配次数,相比其他几种算法有更好的时间性能。     

MOCVD于高温AlN模板上生长的AlGaN材料

A high temperature AlN template was grown on sapphire substrate by metalorganic chemical vapor deposition.AFM results showed that the root mean square of the surface roughness was just 0.11 nm.Optical transmission spectrum and high resolution X-ray diffraction（XRD）characterization both proved the high quality of the AlN template.The XRD（002）rocking curve full width at half maximum（FWHM）was about 53.7 arcsec and（102）FWHM was about 625 arcsec.The densities of screw threading dislocations（TDs）and edge TDs wereestimated to be - 6 × 10^6 cm^-2 and - 4.7 ×10^9 cm^-2. AlGaN of Al composition 80.2% was further grown on the AlN template. The RMS of the surface roughness was about 0.51 nm. XRD reciprocal space mapping was carried out to accurately determine the Al composition and relaxation status in the AlGaN epilayer. The XRD （002） rocking curve FWHM of the AIGaN epilayer was about 140 arcsec and （102） FWHM was about 537 arcsec. The density of screw TDs was estimated to be - 4 × 10^7 cm^-2 and that of edge TDs was - 3.3 × 10^9 cm^-2. These values all prove the high quality of the AlN template and AlGaN epilayer.     

一种基于无线携能通信的MEC与用户匹配算法

多接入边缘计算（Multi-access Edge Computing,MEC）和无线携能通信可有效提高用户的服务质量和体验。在计算、通信和能量等资源的约束条件下,用户匹配是优化MEC任务卸载时系统效用的重要方法。针对无线携能通信的MEC网络结构,综合考虑用户的需求差异性和多元化能量供给,建立了基于计算资源、传输资源和能量资源的系统效用函数;以系统效用最大化为目标,采用基于多维背包理论的多轮拍卖,提出了一种适用于多用户和多网络边缘服务器的用户匹配算法。仿真验证了所提用户匹配算法的有效性与可靠性,结果表明所提匹配算法可优化系统资源配置,有效提高整体性能。     

Efficient Imine Formation by Oxidative Coupling at Low Temperature Catalyzed by High-Surface-Area Mesoporous CeO2 with Exceptional Redox Property

High-surface-area mesoporous CeO₂ (hsmCeO₂) was prepared by a facile organic-template-induced homogeneous precipitation process and showed excellent catalytic activity in imine synthesis in the absence of base from primary alcohols and amines in air atmosphere at low temperature. For comparison, ordinary CeO₂ and hsmCeO₂ after different thermal treatments were also investigated. XRD, N₂ physisorption, UV-Raman, H₂ temperature-programmed reduction, O₂ temperature-programmed desorption, EPR spectroscopy, and X-ray photoelectron spectroscopy were used to unravel the structural and redox properties. The hsmCeO₂ calcined at 400 °C shows the highest specific surface area (158 m² g⁻¹), the highest fraction of surface coordinatively unsaturated Ce³⁺ ions (18.2 %), and the highest concentration of reactive oxygen vacancies (2.4×10¹⁵ spins g⁻¹). In the model reaction of oxidative coupling of benzyl alcohol and aniline, such an exceptional redox property of the hsmCeO₂ catalyst can boost benzylideneaniline formation (2.75 and 5.55 mmol h⁻¹ based on >99 % yield at 60 and 80 °C, respectively) in air with no base additives. It can also work effectively at a temperature of 30 °C and in gram-scale synthesis. These are among the best results for all benchmark ceria catalysts in the literature. Moreover, the hsmCeO₂ catalyst shows a wide scope towards primary alcohols and amines with good to excellent yield of imines. The influence of reaction parameters, the reusability of the catalyst, and the reaction mechanism were investigated.     

Fast pixel-matching for video object segmentation

Video object segmentation, aiming to segment the foreground objects given the annotation of the first frame, has been attracting increasing attentions. Many state-of-the-art approaches have achieved great performance by relying on online model updating or mask-propagation techniques. However, most online models require high computational cost due to model fine-tuning during inference. Most mask-propagation based models are faster but with relatively low performance due to failure to adapt to object appearance variation. In this paper, we are aiming to design a new model to make a good balance between speed and performance. We propose a model, called NPMCA-net, which directly localizes foreground objects based on mask-propagation and non-local technique by matching pixels in reference and target frames. Since we bring in information of both first and previous frames, our network is robust to large object appearance variation, and can better adapt to occlusions. Extensive experiments show that our approach can achieve a new state-of-the-art performance with a fast speed at the same time (86.5% IoU on DAVIS-2016 and 72.2% IoU on DAVIS-2017, with speed of 0.11s per frame) under the same level comparison. Source code is available at https://github.com/siyueyu/NPMCA-net.     

UV nanoimprint using flexible polymer template and substrate

Nano to micro-sized patterns were formed on a flexible polymer substrate using a flexible UV imprint stamp. A 6 in. diameter flexible UV nanoimprint template was fabricated using PVC hot embossing and DLC coating. Using the UV nanoimprint process with the DLC coated PVC template, nano to micro-sized patterns were clearly formed on the flexible PET substrate without a residual layer, due to the antistiction properties and high mechanical hardness of the DLC coating. By depositing a Cr layer on the imprinted resist pattern and lifting it off, Cr metal patterns were fabricated on the PET substrate.     

社交网络智慧搜索研究进展与发展趋势

在以社交网络为代表的Web 2.0时代,搜索需求已经远超出Web 1.0搜索引擎的能力范畴,数据具有模式多态、快速产生、动态交互、碎片化、变化演绎等特点,这些新特点给搜索引擎技术带来了革命性和颠覆性的挑战。归纳了在线社交网络智慧搜索研究进展和技术要点,分析了研究发展过程将面临3个主要问题：在线社交网络中智慧与知识的发掘与推演、用户真实搜索意图的理解与表示以及满足用户真实意图的智慧解答在线响应,并以解决此3个问题为目标讨论了相关技术发展。     

改进自适应加权的局部立体匹配算法

针对Yoon经典自适应加权法在深度不连续处误匹配率较高、边缘细节不突出以及匹配窗口大小凭经验设计不通用等缺陷,提出了一种改进自适应加权的局部立体匹配算法.在经典自适应加权局部立体匹配算法的基础上用高斯分布权值替换了几何空间权值,增加了边缘权值突出深度不连续处的边缘细节,采用赢家通吃原则(Winner Take All,WTA)获取单像素点视差,在目标视图中逐点求取视差,最终获得稠密视差图.实验结果表明,该算法相比于经典自适应加权法,在非遮挡区域错误匹配百分比上下降6％,在深度不连续处的误匹配率降低了5％,边缘细节毛刺明显减少.     

基于ARM和DSP空中运动目标实时跟踪系统研究

对空中目标物体实时跟踪目前主要采用雷达探测的方法,但时常会受到电磁场与电磁波的干扰,降低了探测的准确性,为解决此类问题,提出了一种基于空中视觉实时跟踪系统的解决方案.该方案适用于小范围空中动态物体的跟踪,包括动态目标的差值提取,二值图像的滤波和模板图像的实时更新,以及动态目标的特征提取、云台控制.为提高系统的准确检测性,提出了一种基于改进的动态模板匹配与帧间差分相结合的空中目标检测方法.该方法有效地解决了电磁场干扰、环境变化使动态目标无法捕捉的问题,根据实验表明,该系统能够很好地检测到空中动态物体,具有很好的鲁棒性、实时性和清晰度.     

基于两跳匹配的OFDMA中继网络联合资源分配算法

针对OFDMA中继网络的两跳特性,提出一种基于两跳匹配的中继网络联合资源分配算法。首先根据中继网络的两跳性建立两跳速率匹配模型,然后利用对偶分解理论将中继网络资源分配的主问题分解为：中继选择、子载波分配和功率分配三个主要的子问题并进行联合优化,同时基于中继网络两跳性在子载波分配的过程中考虑两跳子载波配对,以逼近系统最优解。最后为了保证算法的公平性,考虑子载波分配因子约束以优化子载波分配。仿真结果表明：所提算法将中继选择、两跳子载波配对与功率分配联合优化以进一步提升系统吞吐量,同时引入子载波分配因子约束,保证了算法的公平性。