收稿日期: 2016-01-19
修回日期: 2016-11-19
网络出版日期: 2017-05-30
基金资助
国家自然科学基金（No.61371123）资助
Self-Adaption MAC Protocol for M2M Network Based on Q-Learning
Received date: 2016-01-19
Revised date: 2016-11-19
Online published: 2017-05-30
提出了一种针对机器与机器通信网络中数据收集业务的自适应媒体接入控制（medium access control,MAC）层协议，即强化学习混合MAC（Q-learning hybrid MAC,QH-MAC）。这种协议在混合分组MAC（hybrid group MAC,HG-MAC）的基础上增加了基于Q-learning的自适应学习机制，其中心节点可以根据网络状况动态调节竞争时段的时长，因此提高了网络的灵活性和适应性。通过OPNET仿真，将QH-MAC与HG-MAC、多时分址、载波监听多路访问/冲突避免性能进行了比较，结果表明QH-MAC在数据传输速率、能量效率和信道利用率上具有优势。
徐昶, 王聪, 刘灵雅, 李宁 . 基于强化学习的M2M网络自适应媒体接入控制协议[J]. 应用科学学报, 2017 , 35(3) : 317 -325 . DOI: 10.3969/j.issn.0255-8297.2017.03.005
An adaptive medium access control (MAC) protocol Q-learning hybrid MAC (QH-MAC) is proposed for data collection application in machine-to-machine (M2M) networks. The QH-MAC enhances hybrid group MAC (HG-MAC) with an adaptive Q-learning based mechanism, in which the central node dynamically adjusts the COP duration according to the network load. The adaptive learning mechanism improves flexibility and the adaptability of the MAC protocol. QH-MAC is compared with carrier sense multiple access with collision avoidance (CSMA/CA), time division multiple address (TDMA) and HG-MAC by optimized network engineering tool (OPNET) simulations. The results show that HG-MAC is better than the others in terms of data rate, energy efciency and channel utilization.
[1] Akyildiz I F, Su W L, Sankarasubramaniam Y, Cayirci E. A survey on sensor networks[J]. IEEE Communications Magazine, 2002, 40(8):102-114.
[2] Wei Y, Heidemann J, Estrin D. An energy-efcient MAC protocol for wireless sensor networks[C]//International Conference on Computer Communicatims, 2002:1567-1576.
[3] Xu C, Wang C, Liu L Y, Li N. HG-MAC:a energy-efcient protocol for M2M network[C]//International Conference on Information Technology and Management Innovation, 2015:964-973.
[4] Huang P, Xiao L, Mutka M W, Xi N. The evolution of MAC protocols in wireless sensor networks:a survey[J]. IEEE on Communications Surveys & Tutorials, 2013, 15(1):101-120.
[5] Liu Z Z, Elhanany I. RL-MAC:a QoS-aware reinforcement learning based MAC protocol for wireless sensor networks[C]//IEEE Information Conference on Networking Sensing and Control, 2006:768-773.
[6] Misra S, Tiwari V, Obaidat M S. Lacas:learning automata-based congestion avoidance scheme for healthcare wireless sensor networks[J]. IEEE Journal on Selected Areas in Communications, 2009, 27(4):466-479.
[7] Yu C, Mitchell P D, Grace D. Reinforcement learning based ALOHA for multi-hop wireless sensor networks with informed receiving[C]//IET Conference on Wireless Sensor Systems, 2012:1-6.
[8] Zhang W Z, Qin Z Q, Xin J J, Wang L, Zhu M, Sun L, Shu L. UPMAC:a localized load-adaptive MAC protocol for underwater acoustic networks[J]. IEEE Sensors Journal, 2015, 16(11):4110-4118.
[9] Azzem R, Seoud A. TMAC:an automated text mining tool for construction of an annotated corpus to support protein-protein interaction information extraction[C]//Information Computer Technology and Development, 2010:75-79.
[10] Pan L, Wu H, Zeng N F. An efcient and scalable prioritized MAC protocol (PMAC) for backbone communication in wireless sensor networks[C]//International Conference on Sensor Technologies and Applications, 2009:508-513.
[11] Bertsekas B D P. Dynamic programming and optimal control[M]. Nashua:Athena Scientifc, 2010.
[12] Kong L F, Wu J. Dynamic single machine scheduling using Q-learning agent[C]//International Conference on Machine Learning & Cybernetics, 2005, 5(5):3237-3241.
[13] Bianchi G. Performance analysis of the IEEE 802.11 distributed coordination function[J]. IEEE Journal on Selected Areas in Communications, 2000, 18(3):535-547.
/
| 〈 |
|
〉 |
