【摘要】 机会网络是一种由移动对等通信节点组成、节点具有消息存储能力、节点间接触随机出现的网络。现有的机会网络路由协议绝大多数仅考虑针对单个路由目标进行路由决策,不能很好的适应上层应用需求或环境的变化,同时难以提高路由协议的综合性能。面向可加性、可乘性、极值性三种类型的路由目标,通过运用多目标决策理论及蚁群算法,给出了一种机会网络中多目标路由算法(Multiple ObjectiveDecision Making routing,MODM)。该算法中路由决策取决于多项路由指标,能够为不同的应用需求或环境提供不同的路由服务,提高了机会网络下路由算法的综合性能。实验结果表明该算法能够根据不同的需求对多个路由目标进行优化,与传染病、FirstContact等机会网络路由算法相比拥有较好的综合性能。更多还原

【Abstract】 Opportunistic network is a kind of networks composed of mobile peer communication nodes,which provides the message storage capacity and the contact between nodes appears randomly.Existing most routing protocols of opportunistic network only consider single routing objective when they make their routing decisions,which can’t satisfy the requirement from the upper application or environment and can’t improve the comprehensive performance very well.Ant-colony-based algorithm was used to study the multi-objective routing decision-making technology in opportunistic network.It could ensure routing decision depends on a number of routing objectives,and provide different routing service for different application’s or environment’s requirement.The experimental results show that the multi-objective routing decision-making technology in opportunistic network can optimize multiple routing objectives according to different requirements,and the routing protocol performs well comparing with other well-known routing protocol like Epidemic,First Contact and so on.更多还原