物联网学报 ›› 2019, Vol. 3 ›› Issue (2): 35-46.doi: 10.11959/j.issn.2096-3750.2019.00103

• 理论与技术 • 上一篇    下一篇

数据驱动下基于感知频谱的物联网数据传输

李轩衡1,孙怡1,王洁2,张海霞3   

  1. 1 大连理工大学信息与通信工程学院,辽宁 大连116023
    2 大连海事大学信息科学技术学院,辽宁 大连116026
    3 山东大学控制科学与工程学院,山东 济南250002
  • 修回日期:2019-03-20 出版日期:2019-06-30 发布日期:2019-07-17
  • 作者简介:李轩衡(1989- ),男,辽宁沈阳人,博士,大连理工大学信息与通信工程学院讲师,主要研究方向为认知无线网络、动态频谱接入、网络资源协同优化、数据驱动的智能通信等。|孙怡(1964- ),女,辽宁沈阳人,博士,大连理工大学信息与通信工程学院教授,主要研究方向为图像处理。|王洁(1981-),男,河南安阳人,博士,大连海事大学教授、博士生导师,主要研究方向为智能无线感知、无线定位跟踪、人工智能和无线网络。|张海霞(1979- ),女,山东菏泽人,山东大学教授,主要研究方向为大数据、人工智能辅助的无线通信系统关键技术。
  • 基金资助:
    国家自然科学基金资助项目(61801080);大连理工大学基本科研业务费资助项目(DUT18RC(3)012)

Data-driven data transmission of the Internet of things based on sensing spectrum

Xuanheng LI1,Yi SUN1,Jie WANG2,Haixia ZHANG3   

  1. 1 School of Information and Communication Engineering,Dalian University of Technology,Dalian 116023,China
    2 Information Science and Technology College,Dalian Maritime University,Dalian 116026,China
    3 School of Control Science and Engineering,Shandong University,Jinan 250002,China
  • Revised:2019-03-20 Online:2019-06-30 Published:2019-07-17
  • Supported by:
    The National Natural Science Foundation of China(61801080);The Fundamental Research Funds of Dalian University of Technology(DUT18RC(3)012)

摘要:

面向海量数据造成的频谱短缺问题,以共享为解决手段,研究了运营商如何合理利用感知频谱传输数据。考虑物联网设备的局限性,设计了超密集认知异构网络架构,基于流量需求和感知开销设计了最优的接入控制和感知决策方法,以实现网络效用最大化。考虑感知频谱的不确定性,将最优规划方案建模成混合整数随机优化问题,并提出了数据驱动下基于统计特征的概率顽健求解方法,在可用带宽概率分布未知的情况下,统计满足各种服务请求的数据传输要求。

关键词: 频谱共享, 感知频谱, 网络效用, 频谱不确定性, 数据驱动

Abstract:

Facing the problem of spectrum shortage caused by the mass data,in order to share as a solution,how operators use the sensing spectrum reasonably to transmit data was studied.Considering the limitation of Internet of things(IoT) devices,the ultra-dense cognitive heterogeneous network architecture was designed,based on traffic demand and perception cost,an optimal access control and perception decision method was designed to maximize network utility.Considering the uncertainty of the perceived spectrum,the optimal programming scheme was modeled as a mixed integer stochastic optimization problem,and a data-driven probabilistic solution method based on statistical characteristics was proposed.In the case of unknown probability distribution of available bandwidth,data transmission requirements meeting various service requests were counted.

Key words: spectrum sharing, sensing spectrum, network utility, spectrum uncertainty, data-driven

中图分类号: 

No Suggested Reading articles found!