基于贝叶斯网络的复杂系统动态故障树定量分析方法

动态故障树的贝叶斯网络分析方法存在局部组合爆炸和备件门节点失效时间仅能是指数分布的不足.首先,给出动态故障树转换为离散时间贝叶斯网络的方法,该方法使用一个确定性函数来替代条件概率表,避免了局部组合爆炸.然后,根据备件门的失效机理和对应的贝叶斯网络结构特征,解决了备件节点失效时间仅能是指数分布的限制.最后,提出一种基于动态故障树的贝叶斯网络精确推理算法,基于该算法给出了系统失效分布、组件重要度等概率计算.实验结果表明,该方法能有效地分析和评估安全攸关系统的概率特性.     

基于动态矩阵控制算法的网络控制系统的研究

网络控制系统（NCS,Netwroked Control System）是指通过实时网络而构成闭环的反馈控制系统。本文利用动态矩阵控制算法对网络控制系统出现的网络时延等问题进行研究。     

用C8051FXXX系列芯片实现实时时钟功能

当系统时钟使用高频内置振荡器时,辅助振荡器可用来驱动石英作实时时钟（RTC）。系统时钟信号可取自内置和辅助振荡器,改变RTC信号源时不会降低其精度,RTC使用定时器2,它是为给辅助输入信号下降沿计数而设置的。CPO用来把石英振荡波形转换成方波。     

一种城市交通路网实时动态多路口路径导航量子搜索方法

城市交通拥堵日益严重,高效的路径导航方法一直是当前研究的热点和缓解拥堵的主要途径.现有的研究成果主要集中在对单个车辆行驶时间的路径寻优和小规模路网的多车辆均衡化的路径导航,没有实现大规模多车辆多路径的实时动态路径导航.当前研究主要存在以下局限：（1）导航方案评价指标单一,不能充分表示导航方案的优劣;（2）无法实现大规模路网的实时导航.针对这些问题,本文提出一种城市交通路网实时动态多路口路径导航量子搜索方法（A Route Guidance Method based on Quantum Searching for Real-time Dynamic Multi-intersections in Urban Traffic Networks,RGQS）,该方法充分考虑各种因素,实时提供大规模路网的路径导航.本文的实验分别在人工路网和真实路网中验证了RGQS方法相比于对比算法可以使行驶时间减少达到20%.     

基于NTP协议的数字钟设计

针对大型公共场所需放置多台同步时钟的问题,提出了基于网络时间协议（NTP协议）的数字钟设计方案,利用嵌入式网络模块实现网络接口、美信的实时时钟芯片实现时钟的信息存储和更新、奥地利微电子的8通道恒流发光二极管（LED）驱动芯片实现时间显示。该基于NTP协议的数字钟定期向NTP服务器发送请求包,利用服务器的响应包的时间戳可以计算出本地客户机时间,修正客户机时间,使其时间与时间服务器的时间保持一致。该数字钟由于采用NTP网络协议,时间同步于NTP服务器,走时准确,安装方便。     

免费的I/O:改进FPGA时钟分配控制

同步数字系统中的时钟信号（如远程通信中使用的）为系统中的数据传送定义了时间基准。一个时钟分配网络由多个时钟信号组成,由一个点将所有信号分配给需要时钟信号的所有组件。因为时钟信号执行关键的系统功能,很显然应给予更多的关注,不仅在时钟的特性（即偏移和抖动）方面,还有那些组成时钟分配网络的组件。     

基于窄变带宽的实时视频传输算法

根据视频流与实时监视应用的特点,提出一种基于TCP协议的动态双缓冲与双线程视频实时传输算法。在发送方设置视频数据缓存与数据发送缓存,并分别由视频数据输入线程与视频数据发送线程负责管理;视频数据输入线程根据预设的最大等待发送时间与实时计算的网络传输速率,动态调节缓存的大小以及在网络拥塞时有选择性地丢弃视频帧;视频数据发送线程实现视频数据发送与按帧从视频数据缓存获取数据,并实时计算出网络数据传输速率。实验结果表明,本算法能最大限度地利用动态变化的网络带宽,保证视频实时发送至接收方与平稳播放,可有效地应用于窄变带宽网络环境下实时视频监视。     

低功耗SOC的动态时钟管理

介绍了一种系统级设计的时钟管理方案以及功耗管理模块的实现;分析了该方案在实现中可能存在的问题,并给出解决方法。此方案可以显著地降低时钟网络的动态功耗,弥补了现有工具只能在设计后期才能发挥作用的不足,达到了降低整个SOC芯片功耗的目的;引用具体设计项目,说明系统如何动态地调节时钟频率。     

一种支持QoS的拓扑自适应动态组播路由算法

YAM、QoSMIC、DSDMR等一类支持QoS的动态组播路由算法允许组播成员动态地加入和离开，同时为接收方提供多个可选择的组播接入路径，以满足不同应用的QoS需求。但这些算法普遍存在控制信令开销大和结点加入时延长，可扩展性不好等问题。本文在分析这些算法的基础上，提出改进的支持QoS的动态组播路由算法，即拓扑自适应动态组播路由（Topology Adaptive Dynamic Multicast Routing，TADMR）算法。该算法避免了以往算法中大部分盲目的路径搜索，并使结点加入时延不再受限于固定的等待时钟，而与网络拓扑相自适应。性能分析和仿真结果表明，该算法具有较低的控制信令开销和结点加入时延，适用于各种网络规模和群组规模，具有良好的可扩展性。     

基于混合贝叶斯网络的混合系统安全性分析方法

安全关键系统的安全性分析模型本质上是离散和连续失效分布共存的混合模型.传统的故障树和马尔科夫链分析方法仅能处理离散分布或指数分布的系统,难以对混合系统进行安全性分析.针对该问题,以DFT系统安全模型为基础,提出一种基于混合贝叶斯网络的混合系统安全性分析新方法.首先,利用狄拉克函数和单位阶跃函数分别表示动态故障树节点间的确定性关系和时序关系,将动态故障树转换为贝叶斯网络.然后,通过分段多项式来拟合网络节点的不同失效分布,提出一种k段n次多项式混合贝叶斯网络来表示动态故障树.最后,给出该混合贝叶斯网络的推理算法.实验分析表明本方法能有效地进行混合系统安全性分析.     

基于SystemTap的Linux服务器性能分析系统的设计与实现

为了重现Linux服务器运行时所有进程/线程的运行状态、相互间的通信状况及对系统资源如cpu、disk、Network Interface Card等的占用情况,本文通过对动态跟踪工具SystemTap的运行机制的分析,并结合"事件追踪机制",提出了一种Linux平台下高精度、可视化的性能分析系统。该系统采用C/S架构,可由用户自由指定需要监控的事件类型及测量时间,最终以图形化界面向用户呈现分析结果。该系统操作简单,对系统负荷影响可忽略不计,较好的满足的用户的需求。     

使用Delphi进行温度采集系统的设计

介绍了数据采集的产生、发展及方法,并对Delphi编程语言在测控系统中串口通信的方法进行了说明。讨论了在多个任务并发执行环境下多线程的设计方法,以及在线程中进行串行通信的一般过程。最后对程序进行了解释和说明。     

The effect of propagation delay uncertainty on the speed oftime-of-flight digital optoelectronic circuits

Time-of-flight synchronization is a new digital design methodology for optoelectronics that eliminates latches, allowing higher clock rates than alternative timing schemes. Synchronization is accomplished by precisely balancing connection delays. Circuits use pulse-mode signaling and clock gates to restore pulse timing. Many effective pipeline stages are created within combinational logic without extra hardware bounding the stages. Time-of-flight design principles are applicable to packet routing and sorting processors for optical interconnection networks. Circuits are unique because the clock rate is limited primarily by imprecision in propagation delay rather than absolute delay, as in circuits with latches. We develop a general model of delay uncertainty and focus on the effect that static and dynamic uncertainty accumulated over circuit paths has on the minimum feasible clock period. We present a method for traversing the circuit graph representation of a time-of-flight circuit to compute arrival time uncertainty at each pulse interaction point. Arrival time uncertainties give rise to pulse width and overlap constraints. From these constraints we formulate a constrained minimization to find the minimum clock period. We demonstrate our method on circuits implemented with 2×2 electro-optic switches and optical waveguides and find the electronic component of path uncertainty frequently limits speed     

Self‐adaptive clock frequency recovery based on buffer control

With high‐speed packet networks and packet transport networks (PTNs) beginning to support transmissions of high‐speed time division multiplex and multiplexer (TDM) and 3G services, the clock recovery in packet networks becomes significant and challengeable. This paper proposes a self‐adaptive clock frequency recovery based on buffer control to support the clock recovery in packet networks. It includes an estimation part and an amend part, estimates clock frequencies using the dynamic moving average algorithm, amends clock frequencies according to feedbacks, and balances stability, convergence, and system faults. Simulations show that the method has the ability of self‐adaption, avoids the buffer overflow and underflow, and achieves better accuracy, less packet delay, as well as better convergence ability. Copyright © 2011 John Wiley & Sons, Ltd.     

一种低开销的多任务调度模型

为了解决大量线程切换带来的系统开销问题,文章提出了一种多任务调度模型,将实际工作抽象成任务对象,由少量线程调度执行.与传统的多线程模型相比,大大减少了系统中并发运行的线程数目,从而降低了线程切换开销.实践表明,该模型在支持大量任务场合具有较好的效果.     

Power-controlled medium access for ad hoc networks with directional antennas

Directional antennas have the potential to significantly improve the throughput of a wireless ad hoc network. At the same time, energy consumption can be considerably reduced if the network implements per-packet transmission power control. Typical MAC protocols for ad hoc networks (e.g., the IEEE 802.11 Ad Hoc mode) were designed for wireless devices with omnidirectional antennas. When used with directional antennas, such protocols suffer from several medium access problems, including interference from minor lobes and hidden-terminal problems, which prevent full exploitation of the potential of directional antennas. In this paper, we propose a power-controlled MAC protocol for directional antennas that ameliorates these problems. Our protocol allows for dynamic adjustment of the transmission power for both data and clear-to-send (CTS) packets to optimize energy consumption. It provides a mechanism for permitting interference-limited concurrent transmissions and choosing the appropriate tradeoff between throughput and energy consumption. The protocol enables nodes to implement load control in a distributed manner, whereby the total interference in the neighborhood of a receiver is upper-bounded. Simulation results demonstrate that the combined gain from concurrent transmissions using directional antennas and power control results in significant improvement in network throughput and considerable reduction in energy consumption.     

PUMA: From Simultaneous to Parallel for Shared Memory System in Multi-core

In contemporary multi-core systems, memory is shared among a number of concurrent threads. Memory contention and interference are becoming increasingly severe incurring such problems as performance degradation, unfair resource sharing and priority inversion. In this paper, we aim at the challenge of improving performance and fairness for concurrent threads while minimizing energy consumption in main memory. Therefore, we propose PUMA, a novel solution that reduces memory contention and interference by judiciously partitioning threads among cores and allocating each core exclusive memory banks and bandwidth based on thread’s characteristics. Our results demonstrate that PUMA is able to improve both performance and fairness while reducing energy consumption significantly compared to existing memory management approaches.     

一种基于无线多媒体业务的可升降级QoS动态带宽分配与优化策略

无线多媒体网络中的业务包括话音、流媒体、交互类和背景类业务4种,除话音业务外其余3种业务都是可变比特速率业务。对该网络用户资源分配(主要是带宽的分配)若采用传统的固定分配方法,必定陷入资源利用率低下和用户QoS得不到保障的两难境地。该文提出了一种基于无线多媒体业务的动态带宽分配与优化策略,在保证用户QoS的前提下,尽可能提高资源利用率。该文分别从网络和用户两个角度考虑,通过系统容量、业务阻塞率、数据延迟、流媒体的实际传输比和VBR业务综合服务等级等参数,对可升降级QoS无线多媒体网络进行了仿真分析,结果表明,对比传统的网络资源管理策略,该策略大大改善了系统的性能,提高了系统资源利用率。     

基于随路时钟恢复的多源数据光纤传输系统

针对数据互联网络中多源高速并行数据实时传输的问题,提出了一种基于随路时钟恢复的多源数据光纤传输系统,详细介绍了其工作原理和设计思想.系统将现场可编程逻辑门阵列(FPGA)内部高速收发器与专用数字锁相环相结合,给出了随路时钟恢复与数据流量控制的具体实现过程.相比于现有的各类高速并行数据传输解决方案,该系统具备可软件定义的数据接入能力,也能支持更加灵活的随路时钟动态范围.同时,通过设计精简合理的帧结构,推导数据位宽与随路时钟之间的约束关系,有效提高了系统传输带宽.测试结果表明,该系统工作稳定可靠,实时传输效果好,时钟恢复精度可达100 fs,扩展了串并转换与并串转换技术的应用领域.     

Over GHz low-power RF clock distribution for a multiprocessordigital system

Conventional interconnections for digital clock distribution pose a severe power consumption problem for GHz clock distribution due to transmission line losses. Therefore, we have proposed an RF clock distribution (RCD) scheme for high-speed digital applications, in particular a multiprocessor system using global clocking. This paper first reports system power and signal integrity analysis results including skew, jitter, impedance mismatch, and noise for RF clock distribution,especially in the GHz range. Based on this analysis, a novel signal integrity design methodology for RF clock distribution systems is proposed. The clock skew created by process parameter variations are modeled and predicted. The system comprises a RF clock transmitter as a clock generator, an H-tree with junction couplers as a clock distributing network and a RF receiver as a digital clock-recovery module. Flip-chip interconnections for the chip-to-substrate assembly and 0.35 μm TSMC CMOS technology for the RF clock receiver are assumed. EMI analysis for 2 GHz 16-node-board-level RF clock distribution networks is conducted using 3D full-wave EM simulation. Finally, the RCD as a low power and high performance clocking method is demonstrated using HP's Advanced Design System (ADS) simulation, considering microwave frequency interconnection models and process parameter variations. In addition, test vehicles for both 2 GHz 16-node and 5 GHz 64-node board-level RF clock distribution networks were implemented and measured using thin, low-loss, and low permittivity Rogers^Lt; RO3003 high-frequency organic substrate