吉林大学学报(理学版)

• 计算机科学 • 上一篇    下一篇

移动自组织网络Q学习和改进蚁群QoS路由算法

高良诚1,2   

  1. 1. 合肥工业大学 计算机与信息学院, 合肥 230009; 2. 铜陵职业技术学院 信息工程系, 安徽 铜陵 244061
  • 收稿日期:2014-09-29 出版日期:2015-05-26 发布日期:2015-05-21
  • 通讯作者: 高良诚 E-mail:glc912@126.com

QoS Routing Algorithm Based on QLearning and ImprovedAnt Colony in Mobile Ad Hoc Networks

GAO Liangcheng1,2   

  1. 1. School of Computer and Information, Hefei University of Technology, Hefei 230009, China; 2. Department ofInformation Engineering, Tongling Vocational and Technical College, Tongling 244061, Anhui Province, China
  • Received:2014-09-29 Online:2015-05-26 Published:2015-05-21
  • Contact: GAO Liangcheng E-mail:glc912@126.com

摘要:

针对移动自组织网络的QoS路由问题, 提出一种结合Q学习和改进蚁群算法的QoS路由算法, 该算法综合Q学习和蚁群算法的优点, 把Q学习算法的Q值作为蚁群算法的初始信息素, 提高了算法初期的收敛速度, 同时在路径选择时综合考虑节点的能量和负载. 仿真实验表明, 该算法在保证QoS需求的前提下, 增加了路由的有效性和鲁棒性, 降低了能耗, 包投递率、 网络生存时间等指标均较好.

关键词: 移动自组织网络, 服务质量, Q学习, 改进蚁群算法,  , 路由算法

Abstract:

In view of QoS rounting problem in mobile ad hoc networks, the author proposed a QoS routing algorithm integrated with Qlearning and improved ant colony algorithm. The algorithm combines the advantages of Qlearning with those of ant colony algorithm, and it takes Q value of Qlearning algorithm as the initial pheromone of ant colony algorithm, improves the initial convergence speed of the algorithm, at the same time, takes the node energy and load into account in path selection. Simulation results show that on the premise of guaranteeing QoS demand, the algorithm increases the effectiveness and robustness of routing and reduces energy consumption, and besides, packet delivery ratio, network lifetime and other indicators display better performances.

Key words: mobile ad hoc network, quality of service(QoS), Qlearning, improved ant colony algorithm, routing algorithm

中图分类号: 

  • TP393