| 基于强化学习算法的网络拥塞控制 |
| |
| 引用本文： | 王春茹,吴捷,郭红霞.基于强化学习算法的网络拥塞控制[J].计算机工程与应用,2005,41(3):18-20. |
| |
| 作者姓名： | 王春茹 吴捷 郭红霞 |
| |
| 作者单位： | 华南理工大学电力学院,广州,510640;东工业大学自动化学院,广州,510075;华南理工大学电力学院,广州,510640 |
| |
| 基金项目： | 国家自然科学基金资助(编号:60174025) |
| |
| 摘 要： | 论文将强化学习算法应用于网络的拥塞控制中,该网络拥塞控制器可以调节源端发送数据的速率,使网络中可能发生拥塞的节点的缓冲区队列长度逼近给定值,从而避免了拥塞的发生,保证了网络的稳定运行。仿真实验验证了算法的有效性。
|
| 关 键 词： | 强化学习 拥塞控制 缓冲区队列长度 |
| 文章编号： | 1002-8331-(2005)03-0018-03 |
Congestion Control in Networks Based on Reinforcement Learning Algorithm |
| |
| Wang Chunru,Wu Jie,Guo Hongxia.Congestion Control in Networks Based on Reinforcement Learning Algorithm[J].Computer Engineering and Applications,2005,41(3):18-20. |
| |
| Authors: | Wang Chunru Wu Jie Guo Hongxia |
| |
| Affiliation: | Wang Chunru 1,2 Wu Jie 1 Guo Hongxia 11 |
| |
| Abstract: | A controller based on reinforcement learning algorithm for congestion control in networks is proposed in this paper.The controller forces the queue length at the bottle-neck node to the desired steady-state value asymptotically by adjusting the source traffic rate and guarantees the stability of the system.Some simulations are also presented and the result shows the effectiveness of the algorithm. |
| |
| Keywords: | reinforcement learning congestion control buffer queue length |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
