战术通信网业务量建模技术研究及网络性能仿真

战术通信网业务流量模型是对网络规划设计、管理和控制的基础,只有模型精确,才能正确地刻画网络的各项特性.以流量建模技术为主线,对所研究网络的流量特性进行了探讨,并在此基础上提出了战术通信网业务流量的建模方法,同时结合所研究的相关技术对战术通信网性能指标进行了分析.针对在仿真中如何加载业务量分布模型的难题,提出了一种PDF Destination Address模型,利用OPNET PDF概率编辑器及相关控制代码来实现业务量比例分布矩阵的运算,从而完成流量加载.用该方法所加载的业务分布模型与实际业务比例分配关系相符,能有效满足各指挥单元在业务量强度、大小和方向上的需要.     

具有多类索赔到达保险风险模型的破产问题

应用Markov骨架过程的方法和补充变量技巧研究了索赔为多类一般到达的保险风险模型 ,分别得到了破产时间与破产时刻前后资产盈余的联合分布以及破产时间的分布。使得索赔为一般到达的保险风险问题的研究取得了较大的进展。     

基于Stateflow的排队系统仿真方法

论述了排队系统与Stateflow相关概念，利用有限状态机理论与面向对象编程思想，提出了建立基于Stateflow排队系统模型的分析步骤，阐述了应用Stateflow对排队系统进行建模与仿真的方法，使用Stateflow对M／M／1／∞／∞／FIFO类型排队系统进行了具体的建模与仿真。仿真结果表明，基于Stateflow的模型与仿真能够有效地描述排队系统的统计特性，并具有可视化、流程化和层次化的特点。     

不同流量的复杂保障网络抗毁性仿真分析

通过引入流量强度指数和流量分布指数,建立了不同网络流量下的复杂保障网络级联失效抗毁性模型。基于该模型比较分析了无标度网络、随机网络和复杂保障网络在不同流量强度和流量分布下对单个节点的随机失效和故意攻击的抗毁性。结果表明:复杂保障网络的抗毁性随着流量强度的增加急剧下降。此外,流量分布对复杂保障网络的抗毁性也有显著影响。     

备件需求过程的MAP数据拟合算法研究

为了将备件需求过程表示为马尔可夫到达过程(Markovian arrival process, MAP)形式,设计了备件需求到达流的MAP拟合算法.首先,采用EM算法将备件需求到达间隔时间拟合成Hyper-Erlang分布形式,然后利用MAP性质和Bayes公式推导出生成元矩阵的计算公式;随后设计了一个完整的数据拟合流程,并通过实例对算法的效果和效率与已有研究进行了对比.结果表明,该算法在确保拟合效果的同时,能够有效提升拟合效率.     

基于Multi-Agent的工程保障排队系统仿真

针对作战工程保障过程中，工程兵部队保障对象多、作业范围广、时间紧而工兵资源相对不足的特点，利用排队论与Multi-Agent技术建立了工程保障排队系统仿真模型；将任务、保障分队、队列抽象为不同Agent，通过各类Agent间的交互作用仿真工程保障排队现象，提出了一种保障分队作业类型的动态调整算法，采用Repast平台进行实现。实验和分析表明，模型可以真实地模拟作战工程过程中的排队现象，对减少任务等待时间、提高工程保障作业效率有借鉴意义。     

基于优先级队列的多约束无线链路资源调度算法

针对海战场环境中指控数据以先入先出(FIFO)的方式排队发往数据链,当数据链中存在丢失率高和不可达节点时,会导致无线链路带宽资源综合利用率不高的问题,结合仿真模型分析了原因,提出了基于优先级队列的多约束无线链路资源调度算法。通过将该算法与FIFO、优先级队列算法(Priority Queue)进行仿真对比分析,结果表明该算法在存在排队报文时增加了丢失率较低链路上的发送速率,减少了目的不可达报文的发送量,最终降低了重要报文的延迟,提高了数据链的利用率。     

利用ZIP模型估计备品备件需求量

针对使用过程中大部分时间处于贮存状态的武器装备的特点,提出了利用ZIP模型估计武器装备备品备件需求量的方法。即假定需求量为0和需求量服从Poisson分布各占一定比例混合,由此建立了确定备品备件需求量的ZIP模型,并对模型中的参数进行估计。最后,通过模拟数据进行验证,说明了这种方法的可行性。     

随机服务系统简介

服务系统在社会经济各部门中得到广泛应用,它起源于电讯,正和计算机结下不解之缘。由于服务对象(称为顾客)到达系统的间隔时间T和系统服务时间v通常都是随机变量,故常称为随机服务系统。这类系统中常有拥挤排队现象,运筹学中的排队论侧重于其理论研究。运筹它的目的在于,探求顾客利益与服务机构得益之间的协调平衡,力求提高其综合效益。静态优化     

多通道图注意力解耦社交推荐方法

提出具有解耦能力的多通道图注意力社交推荐模型,该模型主要包括深度聚类模块、多通道图注意力聚合模块和评分预测模块。其中,深度聚类模块用于对用户和项目进行分组,并利用聚类结果将用户社交图和用户项目图拆分成多个用户社交子图及用户项目子图,以学习用户兴趣分组及用户对不同类项目的兴趣；多通道图注意力聚合模块学习不同子图对预测结果的注意力；评分预测模块将学习到的用户表示向量和项目表示向量输入多层感知机进行评分预测。在多个真实数据集上的实验结果表明:提出的方法优于其他社交推荐算法。与最新的用于社交推荐的图神经网络方法相比,在Ciao和Epinions数据集上,均方根误差分别降低了2.26%和2.07%,平均绝对误差分别降低了2.58%和3.06%。     

典型 ATM 交换机缓冲系统在类 C/S 业务模式下的队列长度比较

许多应用场合中的通信业务可以用类用户一服务器（C／S）业劳模式来表达,因此有必要研究这种业务下的ATM交换机缓冲存储器策略;本文以计算机仿真的手段评估了四种典型缓冲结构的队列长度概率分布情况,进而比较了类C／S业劳模式下ATM交换机几种典型缓冲系统的性能。     

随机环境中的M/MY/1排队系统

文献［４］讨论了随机环境中的Ｍ／Ｍ／１排队模型，本文提出和讨论随机环境中的Ｍ／Ｍｙ／１排队模型，在统计平衡条件下给出了队长和等待队长的平稳分布以及平均队长和平均等待队长，得到了等待时间和逗留时间分布以及平均等待时间和平均逗留时间。     

Modeling and analysis of vessel casualties resulting from tanker traffic through narrow waterways

In this paper, we present a physics-based stochastic model to investigate vessel casualties resulting from tanker traffic through a narrow waterway. A state-space model is developed to represent the waterway and the location of vessels at a given time. We first determine the distribution of surface current at a given location of the waterway depending on channel geometry, bottom topography, boundary conditions, and the distribution of wind. Then we determine the distribution of the angular drift for a given vessel travelling at a given location of a waterway. Finally, we incorporate the drift probabilities and random arrival of vessels into a Markov chain model. By analyzing the time-dependent and the steady-state probabilities of the Markov chain, we obtain risk measures such as the probability of casualty at a given location and also the expected number of casualties for a given number of vessels arriving per unit time. Analysis of the Markovian model also yields an analytical result that shows that the expected number of casualties is proportional to square of the tanker arrival rate. We present our methodology on an experimental model of a hypothetical narrow waterway. © 1999 John Wiley & Sons, Inc. Naval Reseach Logistics 46: 871–892, 1999     

Individual headstart strategies for combating congestion

This paper examines the process by which a user of a queueing system selects his arrival time to the system to compensate for unpredictable delays in the system if he wishes to complete service at a particular time. Considering the case in which all the system users have already decided on their arrival times to the system and will not change these times, this paper investigates how a new user of this system develops his strategy for selecting his arrival time. The distribution of this customer's arrival time is then obtained for a special case.     

Transit vessel scheduling

A transit vessel traffic scheduling algorithm has been developed to limit the negative effects on cargo volume throughput in two‐way waterways where separation distances between transiting vessels must be maintained and passage restrictions may hold. It runs in time that is polynomial in the number of ships involved in the computation and finds schedules which increase the utilization of waterways. Three examples illustrate its use. The first example is situated in the Sunda Strait where the algorithm is used to enhance the safety of merchant shipping against a terrorist threat. It illustrates important features of the algorithm and demonstrates how it can be used with cross traffic. The second example is situated in the Strait of Istanbul and offers a comparison between the developed algorithm and the transit vessel scheduling algorithm of Ulusçu et al., J Navig 62 (2009), 59–77. This was done using a plausible model of the Strait of Istanbul. The third and last example shows how the algorithm can be used to schedule transit vessel traffic in two‐way waterways with junctions. This feature is especially useful in congested waters with a high risk of collisions like the Inland Sea of Japan. An extreme test case proves that the developed algorithm is a practical algorithm ready for such use. © 2017 Wiley Periodicals, Inc. Naval Research Logistics 64: 225–248, 2017     

EEMD+BiGRU组合模型在短时交通流量预测中的应用

针对城市交通流随机波动性强、数据中含噪声多导致预测精度下降的问题,提出一种基于集合经验模态分解(ensemble empirical mode decomposition,EEMD)和双向门控循环单元(bidirectional gated recurrent unit,BiGRU)的组合交通流量预测模型,有效地提升了短时交通流预测的精度。模型利用EEMD算法对原始数据进行分解,根据分解所得的本征模函数(intrinsic mode function,IMF)分量绘制噪声能量图谱,去除分量中的噪声,并将去噪后的IMF分量作为BiGRU网络的输入进行训练,再将训练所得的结果进行重构加和,得到最终的预测结果。实验结果表明,未舍弃含有噪声的IMF分量进行重构的预测结果,相比于参考文献中提出的EMD+LSTM模型、LSTM模型和EEMD+LSTM模型,其平均绝对百分误差分别优化了42.36%、61.82%和30.95%;舍弃含有噪声的IMF分量后进行重构的预测结果,其平均绝对百分误差相比于将全部IMF分量进行重构优化了56.62%。     

Stochastic call center staffing with uncertain arrival,service and abandonment rates: A Bayesian perspective

In this article, we introduce staffing strategies for the Erlang‐A queuing system in call center operations with uncertain arrival, service, and abandonment rates. In doing so, we model the system rates using gamma distributions that create randomness in operating characteristics used in the optimization formulation. We divide the day into discrete time intervals where a simulation based stochastic programming method is used to determine staffing levels. More specifically, we develop a model to select the optimal number of agents required for a given time interval by minimizing an expected cost function, which consists of agent and abandonment (opportunity) costs, while considering the service quality requirements such as the delay probability. The objective function as well as the constraints in our formulation are random variables. The novelty of our approach is to introduce a solution method for the staffing of an operation where all three system rates (arrival, service, and abandonment) are random variables. We illustrate the use of the proposed model using both real and simulated call center data. In addition, we provide solution comparisons across different formulations, consider a dynamic extension, and discuss sensitivity implications of changing constraint upper bounds as well as prior hyper‐parameters. © 2016 Wiley Periodicals, Inc. Naval Research Logistics 63: 460–478, 2016     

域间流量工程体系结构的研究

随着Internet的快速发展,为了提高网络整体资源的利用率,出现了通过跨域网络性能的优化,达到优化网络资源使用的域间流量工程技术。在分析与评价域间流量工程体系结构研究状况的基础上,提出了面向域间流量工程的RCP体系结构以及今后需要研究的主要问题。     

多参数带约束限制的风险运输网络优化研究

对赋有四个参数（可靠度、时间、费用、容量）的运输网络进行优化分析,给出求解最优路线的算法。