The simulated results show that burst loss probability can be reduced by 1 to 2 orders of magnitude compared with the conventional schemes of Fiber Delay Line in optical switching networks.
仿真结果表明，与传统的光纤延迟线竞争解决方案相比，这一调度算法能够改善交换性能1到2个数量级，是利用光纤延迟线解决光突发交换中竞争问题的一个较佳方案。
Secondly, by researching the switching and delay performances of Ring, Star, Mesh-toms and NFSNet topologies the results exhibit that the loss probability and delay of networks significantly depend on these parameters of network such as connects, network diameter and so on.
其次，采用计算机模拟仿真研究了Ring、Star、Mesh-torus和NFSNet四种典型网络拓扑的交换性能和延时性能，表明网络连接、网络直径等网络参量对网络性能具有很大的影响。
The paper analyzes the present congestion and time-delay situation, discusses the present advantage and limitation of various congestion control mechanism in detail, expounds normal model theory of network, illustrate the theoretical model of network packets arrival and node services, research the modeling and simulating method of simulated mode about network and information data, loop linearity control based on regulation of input rate, loop nonlinearity control based on regulation of input rate, network loop control of several nodes based on neural net, typical active queue management means such as random early detection and adaptive RED, loop control of discard probability of node queue based on the poisson-distributed source, loop control of integrated services discard probability based on multisource share memory, loop control of differentiated services discard probability based on multisource share memory, loop control of network time-delay based on information queued online, and the control mechanisms of concrete control structure and algorithm are proposed.
阐述了计算机网络的一般建模原理，给出了网络信息到达和节点服务的理论模型表述，研究了计算机网络节点及分组信息的仿真模型和仿真方法。提出了基于输入速率的闭环线性调节、基于输出速率的非线性闭环调节、基于神经网络的多节点网络闭环控制机制及算法；对典型的主动队列管理算法，RED和自适应RED算法进行了仿真，提出了基于泊松分布输入的节点队列概率丢弃闭环控制、多信源共享内存的集成服务概率丢弃闭环控制、多信种共享内存的区分服务概率丢弃闭环控制等多种控制机制，并给出了具体的控制结构和控制算法。
An approach similar to spatial smoothing for array signal processing is adopted to recover full rank of convariance matrix of time delays, and the real domain approach to fasten the operation is also in the consideration. At the same time, an approach to mitigate the first time delay errors for a pair of terminal is also presented, which exploits the probability analysis. Several areas are constructed to distinct the estimations, and put the estimations of couples of samples into the distinct area, choose the area which has most estimations, get the last result through the avaraging. The simulation results show that the proposed approach provids high resolution time of arrival estimations.
本文提出了将TLS-ESPRIT算法应用于到达时间估计，并且在小样本的前提下，利用空间平滑的方法将路径的自相关矩阵恢复成满秩，同时将实值运算应用于谱估计算法，加快了运算速度；同时为了减小首径衰落的影响，本文提出了基于概率分析的首径到达时间估计方法，该方法主要选取了数个样本下的前几个估计值，通过划定一定宽度的区间，选取估计值出现频率最高的区间；将该区间中的各个估计值求平均值后，得到最后的首径到达时间估计，实验表明，该方法能够获得较小的估计误差，提高了估计精度。
In the second place, wavelet transform technology is too sensitive to negative-pressure signal singularity, and easy to produce false alarm, so the thesis brings forward to SPRT based on principle of dynamic mass balance principle. In the condition of settled false alarm and missing alarm, the stylebook is least; Aiming at the delay question of SPRT, compensation algorithm method is used, as a result, the speed of leak detection is improved and response time is shortened; When leak quantity is unknown or is estimated enough inaccurate, the thesis brings forward to a method that many SPRT are used to detect leak at one time, and the emulation result proves its probability and validity.
其次针对小波变换技术对压力信号奇异点过于敏感，易产生误报警的问题，基动态质量平衡原理，利用序贯概率比检验方法，在确保泄漏检测的误报率和漏报性能要求的情况下，使检验样本个数最小；文中还对序贯概率比检验方法的延迟问题，采用了相应的补偿算法，从而提高了泄漏检测速度，缩短了响应时；针对泄漏量未知或估计不够精确的问题，提出了利用多个SPRT并行检漏的设，仿真证明了其可能性和有效性。
But, before statistical multiplexing a new type of traffic with existing traffic in the network, it is needed to estimate the performance parameters (such as cell loss probability or delay) of the multiplexer. The accurate estimate of these parameters not only enables us to get the good estimate of the quality of service provided by the ATM network but also provides us with an important parameter that can be used to decide whether or not to accept a connection request so as to avoid long term congestion.
在统计多路复用中，需要对系统的性能参数进行估计，对这些参数的精确估计不仅使我们能够充分了解网络提供的服务质量，而且还可以利用这些参数进行呼叫接入控制，以预防网络拥塞发生。
Firstly, set up the insulation aging and PD measurement method. Ultraviolet lighting was added during the process of PD measurement, which was used to excite photoionization, add the appearance probability of initial electron, and decrease the influence of time delay.
首先确定了尽缘的老化及局部放电测试方法，在局部放电测试过程中加了紫外光照射，通过紫外光激发光电离，增加初始电子的出现概率，减少统计时间延迟的影响。
Firstly, aiming at the defect that shortest path routing algorithms based on Hopfield neural network often fall into undesirable local minima, we propose a shortest path routing algorithm based transiently chaotic neural network, meanwhile, we introduce new energy constraints terms and post-processing technique. The objective of those measures including neural network framework and energy functions is to promote the probability of optimal convergence; Secondly, we discuss the DCLC routing problem, which is a QoS routing problem with NP- completeness characteristic, and propose a genetic QoS routing algorithm. In the algorithm, we try to introduce new coding scheme, corresponding path crossover、mutation operators, and use annealing technique to handle the delay constraint; Lastly, for the delay-bounded multicast routing problem, based on the path sets calculated by the K shortest path algorithm, we research how to apply the extend PBIL algorithm to search the optimal multicast tree.
论文针对基于Hopfield神经网络的最短路径路由算法易陷入局部最小值的缺点，提出一种基于暂态混沌神经网络的最短路径路由算法，引入新的能量函数约束项和事后处理机制，从神经网络结构、能量函数等多方面来提高算法的全局收敛性；论文探讨了具有NP-完全特性的DCLC这一QoS路由问题，提出一种基于遗传算法的QoS路由算法，尝试引入先进的路径编码设计思想和与之相对应的遗传交叉、变异算子，以及如何将模拟退火约束处理技术用来处理时延约束；论文针对时延约束组播路由问题，在采用K条最短路径算法计算得到的路径集合的基础上，就如何利用扩展PBIL进化算法搜索最优组播树展开研究。
In this paper, started with the most elementary probability formula of Fisher matrix for time delay estimation, some properties of the Fisher matrix for time delay, Doppler shift and its hither derivatives joint estimation are exploited by use of Toeplitz matrix chracteristics in passive location systems.
本文从时延估计的Fisher矩阵的最基本概率表达式入手，利用Toerlitz矩阵性质，对无源定位中时延、多普勒及其更高阶导数联合估计的Fisher矩阵的性质进行分析。
In this paper, the delay probability distribution function (pdf) and the approximate expression for the equivalent bandwidth of the real time multimedia flow have been developed with fluid flow method.
本文利用流体流方法计算了实时媒体流在缓冲区中的时延概率分布函数，给出了时延约束下实时媒体流的等效带宽的表达式。
Next, with the means of computer, we simulate the PRN work process, record the average transmission delay and the average packet lost probability.
运用该软件，模拟了分组无线子网的工作过程，并对系统的主要性能参数（网络的平均传输时延和平均丢包率）进行了统计。
By using the disordered resistance to counteract the turn-off delay, the fast shutdown of the Synchronous Transistor will be realized and the probability of the current back flow will be reduced. 6.
该电路利用失调电阻抵消同步管关断延迟，达到了快速关断同步管的目的，有效的减小了电流倒灌现象的发生。
Two novel type of soft-output decoding algorithms for turbo codes with a convolutional interleaver is proposed, based on a decision rule of symbol-by - symbol maximum a posteriori (MAP) probability. Each of them requires only a forward recursion and has lower decoding delay.
基于最大后验概率判决准则，提出了两种只有前向递推的软输出译码算法，它们具有较低的译码延迟，可用于面向连续比特流的turbo码中。
In Polarization Mode Dispersion (PMD) simulation, are analyzed. The experimental results show that low probability events of large Differential Group Delay (DGD) can be resulted by importance sampling simulation and the system configuration is much simpler than by the standard Monte-Carlo method.
实验证明，采用重要抽样的仿真方法可以得到大差分群时延（DGD）值的小概率事件，并且系统配置比用蒙特-卡洛的仿真方法简单得多。
The probability density functions (PDF) of PMD vector's components obey Gauss distribution and the PDF of differential group delay (DGD) time obeys Maxwell distribution;
偏振模色散矢量的各个分量服从高斯分布，差分群时延服从麦克斯韦分布；
Effects of the stochastic delay probability on traffic flow
随机延迟概率对交通流的影响
The probability mass function of queueing delay, the autocorrelation and power spectrum of delay jitter for this periodic cell stream are derived.
得出了该路周期性信元流的排队时延分布、时延抖动的自相关函数和功率谱。
On the APON systems with global FIFO principle based media access control (MAC) protocol, the cell delay property of constant bit rate cell stream is theoretically analyzed. The probability distribution of cell delay is obtained.
针对基于全局逻辑队列先进先出的介质访问控制（MAC）协议，分析了APON系统中一路为固定比特率的信元流的时延特性，得到了信元时延的概率分布结果。
This protocol may save energy at expense of packet delay performance and the probability of queue overflow.
因为牺牲了时延性能，本协议适用于非实时性的数据业务。
By means of queuing analysis and network simulation, the measurability of end-to-end minimum packet delay has been quantificationally analyzed, and it was found that 50 packets was more than enough to measure the minimum delay of a 10 hops long path with very great probability.
以排队分析和网络仿真为手段定量分析了在不同路径长度和流量负载条件下的最小时延可测性，发现约50个探测包即可有很大概率测量到有10跳长路径的最小时延。