通信学报 ›› 2019, Vol. 40 ›› Issue (10): 137-148.doi: 10.11959/j.issn.1000-436x.2019203

• 学术论文 • 上一篇    下一篇

SDM-EON中考虑空闲光路预测的节能算法

熊余1,2,3, 贺进有1,2,3, 王保华1,2,3, 周彬1,2,3   

  1. 1 重庆邮电大学通信与信息工程学院,重庆 400065
    2 重庆高校市级光通信与网络重点实验室,重庆 400065
    3 泛在感知与互联重庆市重点实验室,重庆 400065
  • 修回日期:2019-08-21 出版日期:2019-10-25 发布日期:2019-11-07
  • 作者简介:熊余(1982- ),男,四川资中人,博士,重庆邮电大学研究员,主要研究方向为下一代宽带网络技术、教育大数据技术等。|贺进有(1992- ),男,河南驻马店人,重庆邮电大学硕士生,主要研究方向为空分复用弹性光网络。|王保华(1996- ),男,湖北黄冈人,重庆邮电大学硕士生,主要研究方向为人工智能技术。|周彬(1994- ),男,重庆人,重庆邮电大学硕士生,主要研究方向为空分复用弹性光网络。
  • 基金资助:
    国家自然科学基金资助项目(61401052);国家留学基金委基金资助项目(201608500030);重庆市教委科学技术研究基金资助项目(KJ1400418);重庆市教委科学技术研究基金资助项目(KJ1500445);重庆邮电大学博士启动基金资助项目(A2015-09)

Energy-saving algorithm considering idle light-path prediction in SDM-EON

Yu XIONG1,2,3, Jinyou HE1,2,3, Baohua WANG1,2,3, Bin ZHOU1,2,3   

  1. 1 School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China
    2 Key Laboratory of Optical Communication and Networks in Chongqing,Chongqing 400065,China
    3 Key Laboratory of Ubiquitous Sensing and Networking in Chongqing,Chongqing 400065,China
  • Revised:2019-08-21 Online:2019-10-25 Published:2019-11-07
  • Supported by:
    The National Natural Science Foundation of China(61401052);The Project of the China Scholarship Council(201608500030);The Science and Technology Research Project of Chongqing Municipal Education Commission(KJ1400418);The Science and Technology Research Project of Chongqing Municipal Education Commission(KJ1500445);The Doctoral Start-up Fund of Chongqing University of Posts and Telecommunications(A2015-09)

摘要:

为有效降低空分复用弹性光网络(SDM-EON)中的能耗、阻塞率及多芯光纤中相邻纤芯间串扰,提出了一种考虑空闲光路预测的节能算法。首先,使用极限学习机模型预测网络中各光路的业务量,得出空闲光路集合与各空闲光路的维持时刻阈值;然后,通过预测算法感知空闲光路的实际维持时刻;最后,在实际维持时刻不超过最小维持时刻阈值且芯间串扰低于串扰阈值的空闲光路中,加载能耗最小的空闲光路被选来分配给新业务。仿真结果表明,相对于传统节能算法,在满足SDM-EON串扰限制的前提下,所提算法能够达到更佳的节能效果,且能使阻塞率维持在一个合理的范围。

关键词: 空分复用弹性光网络, 空闲光路预测, 极限学习机, 节能

Abstract:

To effectively reduce the energy consumption,blocking rate and crosstalk between adjacent cores in a multi-core fiber for space division multiplexing elastic optical network (SDM-EON),an energy-saving algorithm considering idle light-path prediction was proposed.The extreme learning machine model was used to predict the traffic volume of each light-path in the network.Thus the idle light-path set and the maintenance time threshold of each idle light-path were obtained.Then,the actual maintenance time of the idle light-path was perceived by the prediction algorithm.Finally,in the light-paths where the actual maintenance time do not exceed the minimum maintenance time threshold and the inter-core crosstalk are lower than the crosstalk threshold,the idle light-path with the least loading energy consumption was selected to carry the new traffic.The simulation results show that compared with the traditional energy-saving algorithm,when the SDM-EON crosstalk limitation is satisfied,the proposed algorithm can lead better energy-saving while maintain the blocking rate at levels compatible.

Key words: SDM-EON, idle light-path prediction, extreme learning machine, energy-saving

中图分类号: 

No Suggested Reading articles found!