| 连续时间Markov决策过程在呼叫接入控制中的应用 |
| |
| 引用本文： | 周亚平,奚宏生,殷保群,唐昊. 连续时间Markov决策过程在呼叫接入控制中的应用[J]. 控制与决策, 2001, 16(Z1): 795-799 |
| |
| 作者姓名： | 周亚平 奚宏生 殷保群 唐昊 |
| |
| 作者单位： | 1. 中国科技大学管理科学系,
2. 中国科技大学自动化系, |
| |
| 基金项目： | 国家自然科学基金项目(69974037);安徽省自然科学基金项目(99043110) |
| |
| 摘 要： | 应用Markov决策过程与性能势相结合的方法,给出了呼叫接入控制的策略优化算法.所得到的最优策略是状态相关的策略,与基于节点已占用带宽决定行动的策略相比,状态相关策略具有更好的性能值,而且该算法具有很快的收敛速度.
|
| 关 键 词： | 呼叫接入控制 Markov决策过程 策略优化 性能势 |
| 文章编号： | 1001-0920(2001)z1-0795-05 |
| 修稿时间： | 2001-06-06 |
Continuous-time Markov Decision Processes to Call Admission Control Problem |
| |
| ZHOU Ya-ping ,XI Hong-sheng ,YIN Bao-qun ,TANG Hao. Continuous-time Markov Decision Processes to Call Admission Control Problem[J]. Control and Decision, 2001, 16(Z1): 795-799 |
| |
| Authors: | ZHOU Ya-ping XI Hong-sheng YIN Bao-qun TANG Hao |
| |
| Affiliation: | ZHOU Ya-ping 1,XI Hong-sheng 2,YIN Bao-qun 2,TANG Hao 2 |
| |
| Abstract: | An algorithm to optimize policy in call admission control is presented by using the method of Markov decision processes combined with performance potentials. The optimal policy obtained by this algorithm is a state-depended policy. Compared to the policy decided by the used bandwidth, this policy has better performance. Furthermore, the convergence speed of this algorithm is very fast. |
| |
| Keywords: | call admission control Markov decision processes policy optimization performance potentials |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
