quotation:		[Copy]
		[Copy]

This Paper:Browse 3823 Download 2900
认知无线Mesh网络自适应多路径算法
章国安,丁晨莉,包志华
0 Fontlarge +\|Default\|Small
(南通大学电子信息学院，江苏南通 226019;东南大学移动通信国家重点实验室，南京 210096)

摘要:

认知无线Mesh网络随着频谱空穴的变化，路径随之改变，给数据传输带来负面影响。提出了一种自适应多路径算法，该算法采用多agent强化学习模型，结合多路径调度策略，通过Q-学习算法更新Q值路由表。在路由变动的情形下，节点能够始终有较优路径传输数据。仿真结果表明，该算法能够减少网络时延，提高网络性能。

关键词: 无线Mesh网络认知无线电强化学习多路径

DOI：

基金项目:国家自然科学基金资助项目(60872002)；江苏省高校自然科学重大基础研究项目(07KJA51007)；江苏省教育厅“青蓝工程”项目；东南大学移动通信国家重点实验室开放研究基金资助项目(W200912)

An Adaptive Multi path Routing Algorithm in CogWMN

ZHANG Guo-an,DING Chen-li,BAO Zhi-hua

(School of Electronics and Information, Nantong University, Nantong 226019, China;National Mobile Communications Research Laboratory, Southeast University, Nanjing 210096, China)

Abstract:

Paths will alter with the changing of spectrum holes,which has negative effe cts on transmissions in Cognitive Wireless Mesh Network(CogWMN). An adaptive mu ltipath algorithm is proposed in this paper. The proposed algorithm adopts multiagen t reinforcement learning model and combines with multipath scheduling strategy t hrough Q-learning algorithm for updated Q-value route tables. Better route can a lways be used for nodes to transmit data when route changes. The simulation resu lts show that the algorithm can reduce network latency and improve network perfo rmance.

Key words: wireless mesh network（WMN） cognitive radio(CR) reinforcement learning multi path