嵌入式操作系统μC／OS-Ⅱ的一种内存管理算法

针对μc／OS-Ⅱ内存管理机制的不足,提出了一种新的内存管理算法．较小的内存分成固定大小的内存块,用位图索引组织;较大的内存用链表组织．实验表明,该方法能较好地提高内存分配速度和利用率,特别是对于内存块大小变化很大的系统．     

一种新的无监督的卫星影像变化检测算法

提出了一种基于主分量分析(PCA)和上下截集模糊Kohonen聚类网络(UDSFKCN)的、无监督的、不同时相的和卫星影像的像素级变化检测新算法.将PCA和UDSFKCN两种方法结合,并将它应用于不同时相的卫星影像的变化检测.该方法结合每个像素的邻域信息,利用PCA,产生每个像素对应的基于邻域信息的特征向量;又将变化区...     

基于自组织特征映射神经网络的聚类分析

在深入研究自组织特征映射（Self-organizing Feature Mapping,SOFM）神经网络的结构和聚类算法的基础上,阐述了SOFM网络的建立方法.以随机二维向量的聚类为例,利用所建立的SOFM网络模型对输入的随机二维向量进行聚类,并着重研究了输出层神经元拓扑结构、训练步数对聚类结果的影响以及在相同拓扑结构条件下,SOFM网络模型的权值向量的调整过程.仿真结果表明：在输出层神经元节点形式为六边型条件下,输出层神经元的个数越多,SOFM网络模型的聚类结果就越准确;在相同的拓扑结构条件下,训练步数越大,SOFM网络聚类结果越准确,但过大的训练步数对于聚类结果的影响甚微.     

基于云模型的高校网络舆情监控系统

设计了一种基于云模型的高校网络舆情监控系统,该系统采用基于立即价值和未来价值综合评价的方式指导网页爬行策略,采用云模型指导爬行方向,同时通过正向和逆向云模型对主题网页进行聚类,并通过查询扩展技术提高网页查全率,取得了很好的实际应用效果.     

一种结合贪心选择和特征加权的高维数据聚类算法

为解决传统聚类算法无法对高维数据聚类的问题,文中提出了一种结合贪心选择和特征加权的TC-Mean shift高维数据聚类算法。通过对一维数据进行聚类,获得一维数据的聚类结果,再通过加权添加维度聚类,最终获得所有维度数据的聚类,实现对高维数据的聚类。测试结果表明,该算法能够准确地对稀疏的高维数据样本进行聚类,能够处理各种维度的数据,具有良好的实际应用价值。     

Vector Quantization Based QoS Classification for Admission Control in CDMA Systems

Optimal power control is of great importance for CDMA systems and it can be controlled to provide the desired quality of service (QoS) to mobile hosts in a cellular radio system. The power levels of all the mobile hosts are determined and constantly tuned in order to achieve the required SINR (signal to interference and noise ratio) which changes dynamically. The SINR of all the K mobiles in a cell can be expressed in the form of a k-dimensional vector. It helps determine the operating point of the system and hence it is constantly monitored and updated due to the variability in the wireless channel conditions and user mobility. We view this continuously changing vector as the motion of a point in a higher dimensional Euclidean space, called the QoS space. We apply vector quantization technique to shrink the infinite-point space to a finite-point space by partitioning the former into N regions such that the points within a region reflect almost similar system performance and are identified by what we call a QoS index. We show how the system operating point can be mapped to one of the QoS indices. The location of the point or the region of operability in the QoS space conveys the system status in terms of the current load and the QoS being delivered. The dynamism in the system's input conditions due to wireless link characteristics and user mobility acts like an opposing force against which the system has to operate. The system reacts to all such changes preventing it from going into a region with an undesirable QoS index. We show how the apriori knowledge of the operating region helps in decision making pertaining to call admission and resource allocation in CDMA systems. Mainak Chatterjee received his Ph.D. from the department of Computer Science and Engineering at the University of Texas at Arlington in 2002. Prior to that, he completed his B.Sc. with Physics (Hons) from the University of Calcutta in 1994 and M.E. in Electrical Communication Engineering from the Indian Institute of Science, Bangalore. He is currently an Assistant Professor in the department of Electrical and Computer Engineering at the University of Central Florida. His research interests include resource management and quality-of-service provisioning in wireless and cellular networks, sensor networks, CDMA data networking, media access control protocols, Internet traffic, and applied game theory. Sajal K. Das is a Professor of Computer Science and Engineering and also the Founding Director of the Center for Research in Wireless Mobility and Networking (CReWMaN) at the University of Texas at Arlington (UTA). His current research interests include resource and mobility management in wireless networks, mobile and pervasive computing, wireless multimedia and QoS provisioning, sensor networks, mobile Internet protocols, distributed processing and grid computing. He has published over 250 research papers, directed numerous funded projects, and holds 5 US patents in wireless mobile networks. He received the Best Paper Awards in ACM MobiCom'99, ICOIN-2001, ACM MSWIM-2000, and ACM/IEEE PADS'97. Dr. Das is also a recipeint of UTA's Outstanding Faculty Research Award in Computer Science in 2001 and 2003, and UTA's College of Engineering Excellence in Research Award in 2003. He serves on the Editorial Boards of IEEE Transactions on Mobile Computing, ACM/Kluwer Wireless Networks, Parallel Processing Letters, Journal of Parallel Algorithms and Applications. He served as General Chair of IEEE PerCom-2004, CIT-2003 and IEEE MASCOTS-2002; General Vice Chair of IEEE PerCom-2003, ACM MobiCom-2000 and HiPC 2000-01; General Chair of ACM WoWMoM 2000-02; Program Chair of IWDC-2002, WoWMoM 1998-99; TPC Vice Chair of ICPADS-2002; and as TPC member of numerous IEEE and ACM conferences. He is the Vice Chair of IEEE TCPP and TCCC. Prior to 1999, Dr. Das was a professor of computer science at Univeristy of North Texas where he twice (1991 and 1997) received the Student Association's Honor Professor Award for best teaching and scholarly research. He received B.Tech. degree in 1983 from Calcutta University, M.S. degree in 1984 from Indian Institute of Science, Bangalore, and PhD degree in 1988 from the University of Central Florida, Orlando, all in Computer Science.     

基于并行小生境技术优化的智能模糊聚类图像分割算法

利用PBM模糊聚类有效性函数以图像特征空间为搜索空间,实现有效性函数的全局寻优,用并行小生境技术解决粒子群(PSO)算法的早收敛问题,优化聚类的全局收敛性能,实现有效聚类数目与聚类中心的并行寻优。通过对遥感图像分割的实验证明,与传统粒子优化群算法的分割结果相比,本文算法拥有更高的有效性且分割效果更优。     

基于Krawtchouk矩不变量和核模糊聚类的快速分形编码

针对基于分类的快速分形编码方法存在着编码速度与解码质量间的矛盾,鉴于Krawtchouk矩不变量具有在仿射变换下保持不变的特性和核模糊聚类在处理非线性问题上的突出优势,本文首次将这两者引入到分形编码中,提出了基于Krawtchouk矩不变量和核模糊聚类的自适应分类快速分形编码方法。首先根据Domain块的方差将其粗分类,再根据Domain块的Krawtchouk矩不变量利用核模糊聚类对Domain块细分类。实验结果表明,与其他基于分类的快速分形编码方法相比,在解码图像质量提高的同时,大大加快了分形编码的速度。     

基于改进K-均值算法的未知雷达信号分选

针对K-均值算法需要事先确定聚类的数目,无法适用于未知雷达信号分选的问题,通过引入脉冲间欧几里德距离和距离阈值TMS2812,完成聚类数目和聚类中心的自动选取,给出一个K-均值的改进算法,改进后的算法既收敛速度快,易于工程化实现,又可自动确定聚类数目和聚类中心。仿真实验表明,该改进算法提高了K-均值算法的适用范围,能够有效适应于未知雷达信号的分选。