强化学习框架下移动自组织网络分步路由算法

移动自组织网络是一种无基础设施、由移动通信节点组成的无线网络,具有高动态特性。传统的路由协议并不能适应节点移动性带来的频繁拓扑变化,简单的洪泛路由也会因开销过大降低网络的性能。针对如何在移动自组织网络中自适应地进行路由选择,提出强化学习框架下的分步路由选择算法。该算法以最小链路总往返时延为目标,基于强化学习进行路由搜寻,在筛选出符合目标需求节点集合的基础上,结合置信度选择路由。在链路变得不可靠时,数据包被广播给筛选出的邻居节点集以提升路由可靠性并降低开销。对提出的算法在分组到达率和路由开销等主要性能指标进行数值仿真分析。仿真结果表明,提出的分步路由算法相比于基于强化学习的智能鲁棒路由,在降低开销的同时,保持着相当的吞吐率。     

军事通信中ad hoc网络路由协议性能分析

移动ad hoc组网灵活,抗毁性很强,将在未来的军事通信中发挥重要作用。因此,对于移动ad hoc网络研究中的首要的路由问题作了探讨,先简要介绍了两种典型的路由协议OLSR(Optimized Link State Routing)和AODV(Ad hoc On-Demand Distance Vector Routing),接着利用网络仿真工具OPNET分析了AODV的协议特点,最后比较了两种路由协议的性能。仿真结果表明AODV路由开销小,但时延大,节点移动过快时存在路由稳定性的问题;OLSR路由开销很大,但时延相对较小。因此在搭建军用adhoc网时,应根据实际使用情形做出选择。     

无线自组织网络跨层攻击检测博弈模型

无线自组织网络中的跨层攻击具有比单层攻击更强的隐蔽性、更好的攻击效果或更低的攻击成本.为了检测无线自组织网络中的跨层攻击,提出了一种基于博弈论的攻击检测模型.由于攻击不可避免地会对各协议层的参数造成影响,因此模型从协议层攻防博弈的角度,建立起相应的策略矩阵和支付矩阵,并通过均衡分析得到该模型的混合策略纳什均衡解.仿真结...     

Currency Warfare and Just War: The Ethics of Targeting Currencies in War

ABSTRACT

Is Currency Warfare defined as, the use of monetary or military force directed against an enemy’s monetary power as part of a military campaign, a just way to fight a war? This article explores the ethics of waging currency warfare against the Just War Tradition’s principles of jus in bello (just conduct in war) and its criteria of discrimination and proportionality. The central argument is that currency warfare is inherently indiscriminate but may be proportionate when policy makers consider the nature of the threat confronted and the targeted currency's level of internationalization, that is, to what degree it is used in foreign transactions or used as a foreign currency reserve. I evaluate this argument against historical cases during the Second World War (1939–1945), the Gulf War (1990–1991), subsequent operations against Saddam Hussein in the early 1990s, and the ongoing campaign against ISIS.