边缘相似度及其在散斑噪声抑制算法比较中的应用

对于滤波算法,噪声抑制能力和边缘保持能力一直是考核其性能的两个重要指标。对于前者,通常都有相应的参数进行表征,而对于后者,通常却只能依靠人眼进行主观评判。根据标准边缘图像和待测边缘图像边缘点位置的相似性,定义了边缘相似度参数,用来表征滤波算法的边缘保持能力。将此参数应用于高斯噪声图像和椒盐噪声图像滤波算法的比较中,取得了预期的效果:此参数值准确地反映了各滤波算法的边缘保持能力。对激光雷达图像而言,散斑噪声滤波算法的边缘保持能力具有非常重要的意义,为此,将边缘相似度用于散斑噪声滤波算法的边缘保持能力比较,结果表明Lee滤波算法和增强Lee滤波算法具有较强的边缘保持能力,而常规的均值滤波和中值滤波边缘保持能力较差。     

Adaptive Transmit Diversity with Quadrant Phase Constraining Feedback

An adaptive transmit diversity scheme with quadrant phase constraining feedback is proposed in this paper. With simple linear operations at both transmitter and receiver, the proposed algorithm can achieve better system performances with only 2M−2 bits of feedback information for systems with M transmit antennas. Theoretical performance bounds of the proposed transmit diversity scheme are derived. Simulation examples and theoretical analyses show that the proposed transmit diversity scheme outperforms not only the conventional open-loop transmit diversity techniques, but also some closed-loop transmit diversity techniques with more information transmitted in the feedback channel.     

A scalable, power-efficient broadcast algorithm for wireless networks

Scalability and power-efficiency are two of the most important design challenges in wireless ad hoc networks. In this paper, we present a scalable, power-efficient broadcast algorithm for wireless ad hoc networks. We first investigate the trade-off between (i) reaching more nodes in a single hop using higher transmission power and (ii) reaching fewer nodes using lower transmission power and relaying messages through multiple hops. Our analysis indicates that multi-hop broadcast is more power-efficient if α ≥ 2.2, where α is the path loss exponent in the power consumption model P(r,α) = c₀⋅ r^α+c₁. Based on the analysis, we then propose Broadcast over Local Spanning Subgraph (BLSS). In BLSS, an underlying topology is first constructed by a localized topology control algorithm, Fault-Tolerant Local Spanning Subgraph (FLSS). FLSS can preserve k-connectivity of the network, where the value of k determines the degree of fault tolerance. Broadcast messages are then simply relayed through the derived topology in a constrained flooding fashion. BLSS is fully localized, scalable, power-efficient, and fault-tolerant. Simulation results show that the performance of BLSS is comparable to that of centralized algorithms. Ning Li received the B.E. and M.E. degrees from Department of Automation, Tsinghua University, Beijing, PR China, in 1998 and 1999, respectively, and the M.S. degree in Computer Engineering from The Ohio State University, Columbus, OH, in 2001 and the Ph.D. degree in Computer Science from University of Illinois at Urbana-Champaign, Urbana, IL. His research interests include design and analysis of wireless mobile ad hoc networks and sensor networks, large-scale network simulation and emulation, and distributed and mobile computing. Jennifer C. Hou received her Ph.D. from The University of Michigan, Ann Arbor, MI in 1993. She is currently a professor in the Department of Computer Science at University of Illinois at Urbana-Champaign, Urbana, IL. Dr. Hou has been supervising several federally and industry funded projects in the areas of network modeling and simualtion, network measurement and diagnostics, enabling communication software for assisted living, and both the theoretical and protocol design aspects of wireless sensor networks. She has published (with her former advisor, students, and colleagues) over 125 papers and book chapters in archived journals and peer-reviewed conferences, and released a truly extensible, reusable, component-based, compositional network simulation and emulation package, J-Sim. She has also served on the TPC of several major networking, real-time, and distributed systems conferences/symposiums, such as IEEE INFOCOM, IEEE ICNP, IEEE ICDCS, IEEE RTSS, IEEE ICC, IEEE Globecome, ACM Mobicom, and ACM Sigmetrics. She is the Technical Program Co-chair of 27th IEEE INFOCOM 2008, First International Wireless Internet Conference 2005, ACM 3rd Information Processing in Sensor Networks (IPSN 2004) and IEEE Real-time Technology and Application Symposium (RTAS 2000). She is severing on the editorial board of IEEE Trans. on Wireless Communications, IEEE Trans. on Parallel and Distributed Systems, IEEE Wireless Communication Magazine, ACM/Kluwer Wireless Networks, Kluwer Computer Networks, and ACM Trans. on Sensor Networks. Dr. Hou was a recipient of an ACM Recognition of Service Award in 2004, a Lumley Research Award from The Ohio State University in 2001, a NSF CAREER award in 1996–2000 and a Women in Science Intiative Award from The University of Wisconsin—Madison in 1993–1995.     

相关衰落信道下的一种层间联合自适应SR-ARQ传输机制

传统自适应传输机制在相关慢衰落信道下有较差的系统性能.针对相关衰落信道,提出了一种层间联合自适应传输机制.利用有限状态马尔可夫链来描述相关衰落信道,同时通过构建一个多状态马尔可夫系统模型来分析系统吞吐性能,并对调制方式和传输分组大小进行优化.另外,推导了系统平均吞吐性能的表达式.仿真结果显示所提出的自适应传输机制在相关衰落信道上能获得更好的系统性能.     

基于请求集与动态令牌的一种对称分布式互斥算法

提出了一种新的分布式互斥算法。该算法通过在基于竞争或请求集的分布式互斥算法中引入动态令牌的概念以及改变某些消息例如应答(reply)、释放(release)等消息的传送方向以及增加各类型消息的信息量将Makawa类算法的消息复杂度从O(3K～5K)降低到O(2K～4K),同时将算法的同步延迟从2T降低至T,并将算法的节点容错能力提高到N?2并保持算法无饥饿,无死锁。通过实际运行和对比,具有较高的使用价值。     

MANET中基于动态拓扑的多路径自适应流量分配算法

移动自主网络中的大多数多路径路由方案几乎没有考虑流量分配和负载均衡问题。显然,无论路径质量好坏给每条路径分配相同的负荷,会大大降低多路径路由机制的性能,最后会大幅增加网络的通信开销和节点的处理开销。围绕这些问题,提出了一个基于动态拓扑的多路径自适应动态流量分配方案,根据路径质量的权重来动态地给多条路径分配数据流量的比例。大多数情况下仅用局部的信息做路由决定,因此大大减少网络开销。在OPNET模拟平台中实现了这个协议,结果表明,该方案对于MANET网络来说的确是一个可行的方法。     

有效提高无线局域网信道可用带宽的MAC机制

802.11 WLAN基础架构BSS（基础服务集）的DCF模式下,有两个因素严重影响可用带宽.一是随着竞争站点数增加,信道竞争开销增大,利用率下降;二是AP（接入点）在竞争信道中并无优势,上下行带宽严重不对称,降低了实际利用率.针对这两个问题提出一种新的MAC机制,首先通过在AP端统计MAC地址来估计竞争站点数,然后根据数学模型计算AP和站点上下行比例控制下的最优初始竞争窗值,通过广播设置以获取信道最大利用率.     

卫星网络中基于最小中断概率的时延受限路由算法

针对卫星网络链路时延长和拓扑时变的特点,给出一种可满足时延限制且在业务持续时间内尽可能不发生中断的时延受限最优寻路(SDCOP)算法,该算法通过预测拓扑链接情况来有效降低链路切换对网络QoS的影响,且具有计算复杂度低,可适应星上设备处理能力的要求等优点。仿真表明算法在计算复杂性、业务阻塞率以及业务中断率方面都优于同类算法。     

使用遗传算法进行微波非线性电路包络仿真

介绍用于分析任意调制信号激励的微波非线性电路包络仿真技术,相应给出了线性器件的伴随模型,并将该方法发展为分析多载波调制信号激励的强非线性电路。此外,本文给出了一种求解非线性谐波平衡方程组或交调波平衡方程组的非数值方法———基于十进制编码的混沌遗传算法(DCGA),融合设计了相应的交叉和变异操作,并将混沌过程引入其中以避免陷入早熟状态,可有效地产生更优解,并加速种群的收敛。最后,给出了运用这种方法的两个计算实例。     

校园计算网格作业自适应调度的研究和实现

由于网格资源具有分布、异构、动态、自治等特性。使得网格作业调度成为网格计算中具有挑战性的任务之一。针对校园高性能计算资源负载不均衡、管理模式不统一的问题，提出了校园计算网格自适应调度系统的解决方案。详细研究了网格作业的白适应调度技术，设计了含有负载均衡能力的资源动态信息收集模型，并给出自适应调度算法。实现了校园范围内高性能计算资源的负载均衡和统一的资源管理。