一种面向弱纹理图像的特征点描述子

现有的大多数特征点提取算法适用于处理纹理丰富的图像,而对于弱纹理图像则无法提取有效的特征点. 对此,提出了多邻域结构张量特征(MNSTF)算法. 基于一系列固定的邻域和图像结构张量,通过表达局部图像的结构和纹理信息,解决了弱纹理和无纹理场景下特征点提取和匹配等相关问题;同时,通过计算邻域之间的相对方向,实现了MNSTF算法特征描述子的旋转不变性. 实验结果表明,MNSTF算法在经过旋转的弱纹理图像测试集上的特征点匹配准确率达到了99.9%以上,验证了其良好的适用性、旋转不变性和鲁棒性.     

基于语义指纹和LCS的文本去重方法

为了解决传统中文文本去重准确率低的问题,本文提出了一种基于语义指纹和LCS的文本去重方法。针对中文文本,预处理后抽取出文本摘要,然后使用tf-idf算法分别得出文本内容特征向量和摘要特征向量,分别将这两个向量作为simhash算法的输入,计算得到文章的内容指纹和摘要指纹。计算两个文本对应的两个指纹的汉明距离,代入本文公式,最终得到这两文本的指纹距离;使用指纹对文本对进行初步筛选,对判定为相似的两个文本使用LCS算法进行进一步对比,避免误判,最终实现中文文本快速去重。实验过程中,通过与LCS算法、simhash算法等多种算法的结果进行对比,可以体现该方法在算法精确度方面的优势,同时,该方法的运行速度优势也能较好地支持大数据量文本的去重操作。     

无线环境下基于多流优先级的并行多路传输

在无线异构网络环境下提出一种基于多流优先级的并行多路传输方案。该方案基于并行多路传输协议CMT-SCTP,并利用了SCTP的多流特性和部分可靠特性。针对数据的重要性和时间敏感度,设计优先级数据调度算法;同时提出基于路径质量的数据分发机制,进一步保证数据的有效传输。在NS-2中的仿真结果表明,该方案能够有效提高数据传输的吞吐量,减少时延,保证较好的数据传输性能。     

基于神经网络的移动机器人路径规划

针对移动机器人在部分环境信息已知情况下的路径规划问题,运用神经网络动态路径最优算法,研究了基于传感器信息的在线路径规划方法.基于扩展卡尔曼滤波的定位方法融合了多个与机器人状态及环境相关的信息,提高了定位精度.首先建立动态的工作空间信息,基于神经网络的路径规划算法完成机器人的路径规划,根据路径点集运动趋向来调节小车移动,完成了导航的任务.对算法的性能和效率进行了分析,实验表明该方法在障碍物的信息未知的情况下,执行速度快.     

File-aware P2P traffic classification: An aid to network management

As P2P dominates Internet traffic in recent years, ISPs are striving to balance between providing the basic networking services for P2P users and properly managing network bandwidth usage. That is, ISPs are required to provide proper bandwidth for each P2P user to get every file to fulfill their provision for communications, while they have to control bandwidth consumption for efficient usage. However, current P2P traffic management strategies are unable to satisfy both requirements. In this paper, our goal is to design a simple and effective scheme for ISPs to moderate the tradeoff. It is achieved by proposing a file-aware P2P traffic classification method that can identify files and the associated flows. The file-level information can lead to more efficient and flexible management strategies on a per-file basis. We offer two alternatives: constraining the per-file bandwidth consumption and the number of per-file concurrent flows. Finally, a real-life trace is measured using our file-aware method from the perspectives of peers and files. The results indicate that ISPs can gain enough opportunities to flexibly choose proper traffic manage parameters according to actual demands.     

Impact of Online Community Structure on Information Propagation: Empirical Analysis and Modeling

Online social networking sites (OSNS), as a popular social media platform, have been developed massively for business and research purposes. In this paper, it investigated the impact of community structure in online social network on information propagation. A SI (Susceptible-Infected) model based on community structure was proposed. In the SI model, the heterogeneity of user’s active time was taken into account. From the results, it was found that the number of links among communities determines the fraction of infected nodes. With the increase of the number of groups G, however, the fraction of infected nodes remains approximately constant. The simulation results will be of great significance: the information will last relatively short for group networks which have either a small or a large number of groups. The results can be useful for optimizing or controlling information, such as propagating rumors in online social networks.     

移动电话资费管制的研究

伴随着中国电信行业,特别是移动电话用户数量的高速发展和有效市场竞争格局的初步建立,移动电话资费的调整、改革与管制等问题成为近几年社会各界关注的焦点.本文从电信管制理论出发,分析和介绍了电信资费管制理论的演进过程,进而对中国电信市场特别是移动电话资费的管制实践进行了系统研究,深入地分析了中国建立的新的移动电话资费管制体系,并对未来移动电话资费管制的发展趋势作了展望.     

信任协商策略的Petri网建模分析方法

对Peer-to-Peer网络环境下信任协商和信任协商策略进行了定义和形式化描述,提出了一种基于Petri网模型的信任协商策略建模分析方法,将资源、证书和资源访问策略映射为Petri网网元进行建模,通过可达分析寻找最优资源解锁序列.和其他信任协商策略建模方法相比,该方法具有较好的性能.     

Evaluation of the pole expressions of nano-scale multistage amplifiers based on equivalent output impedance

Several frequency compensation schemes have been proposed to stabilize multistage amplifiers with negative feedback. The performance of these amplifiers can be analyzed by inspecting their input-output transfer function as representation of their frequency response. With many circuit elements affecting the output response, it is relatively difficult to obtain the real transfer function of multistage amplifiers based on only the original small-signal expressions. Instead, certain techniques such as Miller’s theorem are used to approximate important parameters such as DC gain and dominant pole. These methods are not generally helpful for approximating the nondominant poles which have a critical role on the loop stability of nano-scale amplifiers. With this issue in mind, this work proposes a systematic methodology to achieve the pole expressions of multistage amplifiers with frequency compensation. The key in the proposed technique is to model the equivalent impedance of the compensation loop at the output. The effectiveness of the proposed approach has been verified through comparison between the transfer functions obtained from theory and those transfer functions found in the literature.     

内存数据库在彩铃业务中的应用

传统的磁盘数据库由于I/O瓶颈的限制,愈来愈不能满足实时高性能应用的需求。内存数据库由于能够提供更快的响应速度和更大的事务吞吐量,在电信领域得到愈来愈多的应用。在对内存数据库技术进行研究的基础上,首次将内存数据库应用到彩铃业务中,提出一种改进的彩铃业务数据库结构。测试结果表明,引入内存数据库之后,彩铃应用的性能得到有效提高,并且CPU占用降低,系统能够承载更多的用户。