%0 Journal Article
%T A Reinforcement Learning Based Ant Algorithm for Multiple Constrained QoS Routing Problem<br>基于再励学习蚁群算法的多约束QoS路由方法
%A CHEN Yan
%A YANG Hua-Jiang
%A SHEN Lin-Cheng
%A <br>陈岩
%A 杨华江
%A 沈林成
%J 计算机科学
%D 2007
%I 
%X This paper discusses the multiple constrained QoS routing problem. Firstly, a mathematical model based on fuzzy judgment is presented, which realizes the optimization of multiple constraint of QoS. Then an Ant algorithm is proposed to solve the problem. An efficient reinforcement learning mechanism, which improves the pheromone according to the reinforcement signal generated from the judgement of the routes, is introduced to the algorithm, so that the algorithm can converge to the approximate global best solution fast. Simulation results demonstrate that the algorithm can effectively and fast generate a route which can mostly satisfy the QoS constraints of operations.
%K Multiple constrained QoS
%K Fuzzy judgement
%K Network routing
%K Reinforcement learning
%K Ant algorithm<br>多约束QoS
%K 模糊评判
%K 网络路由
%K 再励学习
%K 蚁群算法
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=5B3AB970F71A803DEACDC0559115BFCF0A068CD97DD29835&cid=8240383F08CE46C8B05036380D75B607&jid=64A12D73428C8B8DBFB978D04DFEB3C1&aid=3488A9CCB8A2C5CC0BFE85E663286A67&yid=A732AF04DDA03BB3&vid=339D79302DF62549&iid=94C357A881DFC066&sid=C5154311167311FE&eid=DB817633AA4F79B9&journal_id=1002-137X&journal_name=计算机科学&referenced_num=6&reference_num=9