光互连网络中资源预留的多级光互连仲裁机制

基于资源预留策略提出一种多级光互连仲裁机制,通过将网络分级实现快速、高效的仲裁。多优先级数据缓存队列的传输节点设计,提供了不同类型流量的差异化传输;通过预约式两级仲裁机制,实现网络的完全公平与100%的高吞吐率。设计并对快速仲裁通道进行了合理布局,极大地缩短了仲裁延迟。仿真结果表明:采用基于资源预留的分级仲裁策略,在多种流量模式下所有节点均获得公平的服务。与Feather Weight相比,分级仲裁策略吞吐率提高17%;与2-pass相比,仲裁延迟减少15%,同时,功耗减少5%。     

蜂窝网络中基于干扰控制的D2D通信联合资源分配方案

在蜂窝网络中部署D2D(Device-to-Device)通信能够有效提升频谱利用率,降低基站负载,但D2D用户与蜂窝用户共享无线信道时会产生信号干扰。提出了一种联合资源分配算法,通过综合考虑信道分配对网络中已有的蜂窝用户和D2D用户的信号干扰,并在小区范围内寻找具有最小干扰值的信道资源分配给用户,以实现有效的干扰控制。仿真结果显示:联合资源分配能够提升D2D链路、蜂窝链路的信噪比及系统总吞吐量,使得蜂窝网络的整体性能优于独立资源分配。     

基于Multi-agent协商的多干扰机资源管理

多干扰机协同是组网雷达对抗的一种重要方法。针对多干扰机协同中面临的数据传输率、时效性等因素所引起的干扰资源管理问题,建立了基于中心控制协商和公约协商的干扰资源管理模型,并给出了相应的协同算法与仿真。仿真结果表明：两种方法在组网雷达对抗中均能取得良好的干扰效果及较低的漏威胁目标信号脉冲比例。其中,基于公约的Multi?agent协商在组网干扰资源管理方面具有更好的可行性,其效果优于基于中心控制协商的方法。     

对新疆高校校园网建设的思考与探索

随着新疆高饺的校园网建设逐渐由硬件基础建设阶段转入软件和资源建设阶段,本文就新疆高校校园网建设的现状及出现的一些问题做了探讨和分析,并对如何加强高校校园网建设,充实网络教学资源,强化校园网知识培训,提高师生应用能力等一系列问题提出了建议,以期为新疆高校的信息化建设提供一些借鉴。     

知识的通信资源管理决策支持系统

专家系统可以模拟专家分析和解决问题的过程进行推理,最终给出合理的方案.将人工智能技术引入通信资源管理,提出了基于知识的通信资源管理方法,讨论了基于该方法的通信资源管理决策支持系统体系结构、知识库资源优化和推理机制.     

主动队列管理机制中PI算法的一种参数配置方法

研究了主动队列管理机制中PI控制器的参数配置方法。AQM中PI算法默认的参数以及一些文献提出的参数配置方法主要是从稳定性考虑,没有保证系统的动态性能指标。在控制论中,"二阶最优模型"是具有较好的动态性能和稳态性能的二阶系统。提出了一种PI控制器的参数配置方法,可以使TCP/AQM控制模型的开环传递函数具有"二阶最优模型"的形式。NS仿真结果表明采用该方法配置的参数可以提高PI控制器的动态性能并保证系统的稳态性能。最后指出PI控制器参数配置对网络状态的敏感性,以及今后的研究方向。     

一个输入依赖于系统状态的排队模型的瞬时队长分布

探讨了一个有如下特征的排队系统:该系统的第m个到达间隔αm-αm-1(m≥1)的分布依赖系统在时刻αm-1拥有的顾客数目。应用补充变量方法,这个系统的瞬时队长分布的积分表示被得到。在各到达间隔分布和诸服务时间分布均具有密度函数的条件下,这个积分表示的被积项能够递归地求取。     

An efficient metaheuristic for integrated scheduling and staffing IT projects based on a generalized minimum cost flow network

Scheduling IT projects and assigning the project work to human resources are an important and common tasks in almost any IT service company. It is particularly complex because human resources usually have multiple skills. Up to now only little work has considered IT‐specific properties of the project structure and human resources. In this article, we present an optimization model that simultaneously schedules the activities of multiple IT projects with serial network structures and assigns the project work to multiskilled internal and external human resources with different efficiencies. The goal is to minimize costs. We introduce a metaheuristic that decomposes the problem into a binary scheduling problem and a continuous staffing problem where the latter is solved efficiently by exploiting its underlying network structure. For comparison, we solve the mixed–binary linear program with a state–of–the–art commercial solver. The impacts of problem parameters on computation time and solution gaps between the metaheuristic and the solver are assessed in an experimental study. Our results show that the metaheuristic provides very favorable results in considerable less time than the solver for midsize problems. For larger problems, it shows a similar performance while the solver fails to return feasible solutions. © 2012 Wiley Periodicals, Inc. Naval Research Logistics 59: 111–127, 2012     

基于时间窗的宽带阵列多目标干扰资源调度

随着宽带数字阵列技术和有源干扰技术的不断发展和成熟,舰载有源干扰设备的未来发展方向将是宽带数字阵列有源干扰系统。针对多目标干扰问题,提出了目标干扰时间窗概念和两种时间窗的处理方法,并建立干扰资源调度模型,通过仿真计算和对比分析,得出基于模糊时间窗的资源调度模型性能优于基于硬时间窗的资源调度模型,能够对干扰资源进行合理的调度,提高宽带阵列多目标干扰能力。     

Inspection games with local and global allocation bounds

This article discusses a two‐player noncooperative nonzero‐sum inspection game. There are multiple sites that are subject to potential inspection by the first player (an inspector). The second player (potentially a violator) has to choose a vector of violation probabilities over the sites, so that the sum of these probabilities do not exceed one. An efficient method is introduced to compute all Nash equilibria parametrically in the amount of resource that is available to the inspector. Sensitivity analysis reveals nonmonotonicity of the equilibrium utility of the inspector, considered as a function of the amount of resource that is available to it; a phenomenon which is a variant of the well‐known Braess paradox. © 2013 Wiley Periodicals, Inc. Naval Research Logistics, 2013