一个基于增强学习算法的路由模型

由于Internet的不断发展，现有的路由算法为适应不同的网络要求，从一开始的RIP、OSPF、BGP等几种，衍生出很多新的适用于特殊网络的路由协议。本文提出一种基于增强学习算法的路由模型。将每个路由节点看作一个Agent，利用增强学习算法的思想使得每个节点在不了解网络拓扑结构的情况下从向邻居转发的概率获得网络的信息，这样路由节点可以选择一个较优的转发方向。同时，节点能对网络的拥塞等情况作出调整。该模型为一些具体网络的路由协议，特别是QoS类路由算法提出了一个新的路由思想。

A Routing Model Based on Reinforcement Learning Algorithm

With the development of internet, the routing algorithms have been improved for some special demands. Based on some former algorithms such as RIP,OSPF,BGP, some new algorithms have been given. This paper approaches a routing model which is based on reinforcement learning algorithms. Every routing note is treated as an Agent. Using the idea of reinforcement learning algorithms, the notes can gain some information of the net from the probability of forwarding, though it does not know the topology of the net. So the note can choose a better forwarding path. Meanwhile it can make some changes, when the block of net is occurred. This model also gives a new idea for routing algorithms which used in some special field such as QoS Routing.