Hierarchical Simulated Annealing-Reinforcement Learning Algorithm for ABR Traffic Control of ATM Networks
For the congestion problems in asynchronous transfer mode (ATM) networks, a hybrid intelligent hierarchical controller based on reinforcement learning (RL) and simulated annealing (SA) algorithm is proposed. The RL algorithm shows the particular superiority in ATM networks, which is independent of the mathematic model and just needs simple fuzzy information in learning process. The proposed hierarchical architecture is able to generate the knowledge base. The SA algorithm is a powerful way to solve hard combinatorial optimization problems, which is used to adjust the parameters of the controller. With the advantages of the two algorithms, the proposed controller forces the queue size in the multiplexer buffer to the desired value by adjusting the source transmission rate of the available bit rate (ABR) service. Simulation results show that the proposed controller can effectively avoid the occurrence of congestion.
Xin Li Yuanwei Jing
School of Information Science and Engineering Northeastern University Shenyang, Liaoning 110004 P.R.China
国际会议
南宁
英文
2007-07-20（万方平台首次上网日期，不代表论文的发表时间）