基于机器学习的大规模机器通信随机多址接入技术研究开题报告

本课题的现状及发展趋势：

在IoT网络中，由于出现不规律且大规模的MTC请求，现如今的随机接入协议遇到随机多址接入信道阻塞的问题，对信息传输十分不利。针对这一问题，近年来，人们提出了用机器学习来改善信道拥塞状况。一些建议集中在运用强化学习来缓解状况，提高参数指标，解决大规模M2M通信拥塞问题。

本课题的价值：

本课题实现将 机器学习与随机多址接入协议相结合，实现了基于强化学习的随机多址接入算法的设计，通过MATLAB仿真了性能，并通过重新设计面向多址接入时延优化的强化学习系统，提高大规模机器通信随机接入的性能。

参考文献：

[1]N. Jiang, Y. Dengand A. Nallanathan, "Traffic Prediction and Random Access ControlOptimization: Learning and Non-Learning-Based Approaches," in IEEECommunications Magazine, vol. 59, no. 3, pp. 16-22, March 2021, doi:10.1109/MCOM.001.2000099.

[2]B. Zhao, G. Ren, X.Dong and H. Zhang, "Distributed Q-Learning Based Joint Relay Selectionand Access Control Scheme for IoT-Oriented Satellite Terrestrial RelayNetworks," in IEEE Communications Letters, vol. 25, no. 6, pp.1901-1905, June 2021, doi: 10.1109/LCOMM.2021.3061717.

[3]S. K. Sharma and X.Wang, "Toward Massive Machine Type Communications in Ultra-DenseCellular IoT Networks: Current Issues and Machine Learning-AssistedSolutions," in IEEE Communications Surveys Tutorials, vol. 22, no.1, pp. 426-471, Firstquarter 2020, doi: 10.1109/COMST.2019.2916177.

[4]D. -D. Tran, S. K.Sharma and S. Chatzinotas, "BLER-based Adaptive Q-learning for EfficientRandom Access in NOMA-based mMTC Networks," 2021 IEEE 93rd VehicularTechnology Conference (VTC2021-Spring), 2021, pp. 1-5, doi: 10.1109/VTC2021-Spring51267.2021.9448787.

[5]X. Hu and J. Sun,"Interference Analysis and Resource Allocation of Burst Scenario inMassive Machine-Type Communications," 2018 IEEE 18th InternationalConference on Communication Technology (ICCT), 2018, pp. 822-826, doi: 10.1109/ICCT.2018.8600067.

[6]C. Di, B. Zhang, Q.Liang, S. Li and Y. Guo, "Learning Automata-Based Access Class BarringScheme for Massive Random Access in Machine-to-Machine Communications,"in IEEE Internet of Things Journal, vol. 6, no. 4, pp. 6007-6017, Aug. 2019,doi: 10.1109/JIOT.2018.2867937.

[7]J. Youn, J. Park,S. Kim, C. You and S. Cho, "A Study of Random Access for MassiveMachine-type Communications: Limitations and Solutions," 2021 IEEERegion 10 Symposium (TENSYMP), 2021, pp. 1-4, doi: 10.1109/TENSYMP52854.2021.9550999.

[8]G. J. Sutton etal., "Enabling Technologies for Ultra-Reliable and Low LatencyCommunications: From PHY and MAC Layer Perspectives," in IEEECommunications Surveys Tutorials, vol. 21, no. 3, pp. 2488-2524,thirdquarter 2019, doi: 10.1109/COMST.2019.2897800.

[9]D. -D. Tran, S. K.Sharma, S. Chatzinotas and I. Woungang, "Q-Learning-Based SCMA forEfficient Random Access in mMTC Networks With Short Packets," 2021 IEEE32nd Annual International Symposium on Personal, Indoor and Mobile RadioCommunications (PIMRC), 2021, pp. 1334-1338, doi:10.1109/PIMRC50174.2021.9569713.

[10] S. K. Sharma and X. Wang, "CollaborativeDistributed Q-Learning for RACH Congestion Minimization in Cellular IoTNetworks," in IEEE Communications Letters, vol. 23, no. 4, pp. 600-603,April 2019, doi: 10.1109/LCOMM.2019.2896929.

[11] H. R. Mazandarani and S. Khorsandi, "Preamble Reusefor Massive Machine-Type Communications in LTE Networks," ElectricalEngineering (ICEE), Iranian Conference on, 2018, pp. 1652-1657, doi:10.1109/ICEE.2018.8472433.

[12]范平志,李里,陈欢,程高峰,杨林杰,汤小波.面向大规模物联网的随机接入：现状、挑战与机遇[J].通信学报, 2021, 42(04): 1-21.

[13]李恒民,张晶,高宏旭.一种基于动态优先级的NB-IoT随机接入算法[J].南京邮电大学学报(自然科学版),2018,38(06):42-47.DOI:10.14132/j.cnki.1673-5439.2018.06.007.

[14]刘冬雪,王聪,李宁,谢威.一种提高大规模M2M通信接入成功率的分层接入方法[J].通信技术,2017,50(08):1683-1690.