基于Sarsa算法和蚁群优化的监测网络路由控制设计

Design of Routing Control for Monitoring Network Based on Sarsa Algorithm and Ant Colony Optimization

摘要:

为了克服已有监测网络路由控制方法所具有的传输路径长和能量消耗高的缺点,设计了一种基于Sarsa算法和蚁群优化算法混合模型的路由控制策略；首先,建立了以MA(Mobile Agent)为基础的路由控制模型,并以访问时间和能耗消耗为基础设计了目标函数,然后,基于改进的蚁群算法设计了一种从数据发送节点到Sink节点的路由设计方法,采用节点之间的距离作为冗余度标准获取合理的下一跳邻居节点集,并采用Sarsa算法来对各路径的信息素进行更新,最后,对基于Sarsa算法和蚁群优化算法混合模型的路由控制算法进行了描述；在NS2环境下进行仿真实验,结果证明文中方法能有效实现网络路由控制,最优路由的获取过程的总仿真时间为300ms,是一种适用于监测网络路由控制的有效方法。 

Abstract:

In order to conquer the defects of the given monitoring network having the defects of transporting long distance and consuming much energy, an algorithm based on Sarsa algorithm and ant colony optimization is proposed. Firstly, the routing control model is built on the basis of MA and the goal function is designed based on accessing time and energy consumption, then the routing is designed based on ant colony algorithm from the transmitting node to the Sink node, the distance between nodes is used as the redundant degree to get the next hop neighbor node set, and the Sarsa algorithm is used to renew the sophomore of the routing. Finally, the routing control algorithm based on Sarsa algorithm and ant colony optimization is described. The experiment is implemented in the NS2 environment, the result shows the method in this paper can realize the network routing control, the sum simulation time is 300ms for the optimal routing obtaining, therefore, it is an effective method of routing control in monitoring network.