您的位置:山东大学 -> 科技期刊社 -> 《山东大学学报(工学版)》

山东大学学报 (工学版) ›› 2019, Vol. 49 ›› Issue (2): 88-95.doi: 10.6040/j.issn.1672-3961.0.2018.342

• 机器学习与数据挖掘 • 上一篇    下一篇

基于MGU的大规模IP骨干网络实时流量预测

郭芳1(),陈蕾1,2,3,杨子文1   

  1. 1. 南京邮电大学计算机学院,江苏 南京 210023
    2. 江苏省无线传感网高技术研究重点实验室,江苏 南京 210023
    3. 南京航空航天大学计算机科学与技术学院,江苏 南京 210016
  • 收稿日期:2018-08-13 出版日期:2019-04-20 发布日期:2019-04-19
  • 作者简介:郭芳(1992—),女,云南建水人,硕士研究生,主要研究方向为机器学习与网络流量预测. E-mail:lighter_around@163.com
  • 基金资助:
    江苏省自然科学基金(BK20161516);中国博士后科学基金(2015M581794);国家自然科学基金(61872190)

Real-time traffic prediction based on MGU for large-scale IP backbone networks

Fang GUO1(),Lei CHEN1,2,3,Ziwen YANG1   

  1. 1. School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing 210023, Jiangsu, China
    2. Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks, Nanjing 210023, Jiangsu, China
    3. College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, Jiangsu, China
  • Received:2018-08-13 Online:2019-04-20 Published:2019-04-19
  • Supported by:
    江苏省自然科学基金(BK20161516);中国博士后科学基金(2015M581794);国家自然科学基金(61872190)

摘要:

为克服长短时记忆网络(long short-term memory, LSTM)计算成本相当大的弊端,提出基于最小门控单元(minimal gated unit, MGU)的大规模IP骨干网络实时流量预测方法。试验结果表明,与基于LSTM的流量预测方法相比,该方法以较少的模型训练时间获得了相当甚至略优的流量预测性能,在流量预测精度和实时性方面也优于已有的前馈神经网络(feed forward neural network, FFNN)和门控循环单元神经网络(gated recurrent unit, GRU)方法。

关键词: 网络流量预测, 大规模IP骨干网, 循环神经网络, LSTM, MGU

Abstract:

In order to overcome the shortcomings of long short-term memory (LSTM) computing cost, a real-time traffic prediction method based on minimum gated unit (MGU) for large-scale IP backbone networks was proposed. The experimental results showed that compared with the LSTM-based traffic prediction method, the proposed method achieved fairly or even better traffic prediction performance with less model training time, meanwhile it outperformed the most advanced feed forward neural network (FFNN), LSTM and gated recurrent unit(GRU) in terms of prediction accuracy and real-time performance.

Key words: network traffic prediction, large-scale IP backbone networks, recurrent neural network, long short-term memory, minimal gated units

中图分类号: 

  • TP393

图1

LSTM数据流及操作"

图2

GRU数据流及操作"

图3

MGU数据流及操作"

图4

创建训练数据的滑动窗口"

图5

基于MGU单元网络流量预测模型"

图6

基于不同模型,第53对OD对流量预测结果"

表1

基于不同单元模型的总参数与平均MAE值"

单元名称 #参数 平均MAE(103)
FFNN 10 689 2.47
LSTM 49 985 2.10
GRU 37 505 1.95
MGU 25 025 1.84

图7

MGU模型与FFNN模型预测网络流量的MAE值分布图"

图8

MGU模型与LSTM模型预测网络流量的MAE值分布图"

图9

MGU模型与GRU模型预测网络流量的MAE值分布图"

1 QU H , MA W T , ZHAO J H , et al. Prediction method for network traffic based on maximum correntropy criterion[J]. China Communications, 2013, 10 (1): 134- 145.
doi: 10.1109/CC.2013.6457536
2 田中大, 李树江, 王艳红, 等. 高斯过程回归补偿ARIMA的网络流量预测[J]. 北京邮电大学学报, 2017, 40 (6): 65- 73.
TIAN Zhongda , LI Shujiang , WANG Yanhong , et al. Network traffic prediction based on ARIMA with gaussian process regression compensation[J]. Journal of Beijing University of Posts and Telecommunications, 2017, 40 (6): 65- 73.
3 AZZOUNI A, PUJOLLE G. A long short-term memory recurrent neural network framework for network traffic matrix prediction[EB/OL]. (2017-06-08)[2018-04-15]. https://arxiv.org/abs/1705.05690.
4 LANER M , SVOBODA P , RUPP M . Parsimonious fitting of long-range dependent network traffic using A RMA models[J]. IEEE Communications Letters, 2013, 17 (12): 2368- 2371.
doi: 10.1109/LCOMM.2013.102613.131853
5 YADAV R K , BALAKRISHNAN M . Comparative evaluation of ARIMA and ANFIS for modeling of wireless network traffic time series[J]. Eurasip Journal on Wireless Communications & Networking, 2014, 2014 (1): 8- 15.
6 KATRIS C , DASKALAKI S . Comparing forecasting approaches for Internet traffic[J]. Expert Systems with Applications, 2015, 42 (21): 8172- 8183.
doi: 10.1016/j.eswa.2015.06.029
7 NIE L S , JIANG D D , GUO L , et al. Traffic matrix prediction and estimation based on deep learning in large-scale IP backbone networks[J]. Journal of Network & Computer Applications, 2016, 76 (C): 16- 22.
8 LIANG Y , QIU L . Network traffic prediction based on SVR improved by chaos theory and ant colony optimization[J]. International Journal of Future Generation Communication & Networking, 2015, 8 (1): 484- 488.
9 Xiang C , Qu P , Qu X . Network traffic prediction based on MK-SVR[J]. Journal of Information & Computational Science, 2015, 12 (8): 3185- 3197.
10 HONG W C . Application of seasonal SVR with chaotic immune algorithm in traffic flow forecasting[J]. Neural Computing and Applications, 2012, 21 (3): 583- 593.
doi: 10.1007/s00521-010-0456-7
11 LV Y , DUAN Y , KANG W , et al. Traffic flow prediction with big data: a deep learning approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 16 (2): 865- 873.
12 江露琪, 孙文胜. 基于改进的BP神经网络的网络流量预测模型[J]. 通信技术, 2017, 50 (1): 68- 73.
JIANG Luqi , SUN Wensheng . Research and implementation of network traffic prediction based on modified BP neural network[J]. Communication Technology, 2017, 50 (1): 68- 73.
13 SHAO H X, SOONG B H. Traffic flow prediction with Long Short-Term Memory Networks (LSTMs)[C]// Region 10 conference (TENCON 2016). Singapore: IEEE, 2017: 2986-2989.
14 LUO X , ZHOU W , WANG W , et al. Attention-based relation extraction with bidirectional gated recurrent unit and highway network in the analysis of geological data[J]. IEEE Access, 2018, 27 (6): 5705- 5715.
15 FU R, ZHANG Z, LI L. Using LSTM and GRU neural network methods for traffic flow prediction[C]//In Proceedings of the Youth Academic Annual Conference of Chinese Association of Automation (YAC). Wuhan: IEEE, 2016: 324-328.
16 ZHOU G B , WU J , ZHANG C L , et al. Minimal gated unit for recurrent neural networks[J]. International Journal of Automation and Computing, 2016, 13 (3): 226- 234.
doi: 10.1007/s11633-016-1006-2
17 HOCHREITER S , SCHMIDHUBER J . Long short-term memory[J]. Neural Computation, 1997, 9 (8): 1735- 1780.
doi: 10.1162/neco.1997.9.8.1735
18 BENGIO Y , SIMARD P , FRASCONI P . Learning long-term dependencies with gradient descent is difficult[J]. IEEE transactions on neural networks, 1994, 5 (2): 157- 166.
doi: 10.1109/72.279181
19 CHUNG J, GULCEHRE C, CHO K H, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[EB/OL]. (2014-12-11)[2018-04-15]. https://arxiv.org/abs/1412.3555.
20 LAKHINA A , PAPAGIANNAKI K , CROVELLA M , et al. Structural analysis of network traffic flows[J]. Acm Sigmetrics Performance Evaluation Review, 2004, 32 (1): 61- 72.
doi: 10.1145/1012888
21 JIANG D D , WANG X , GUO L , et al. Accurate estimation of large-scale IP traffic matrix[J]. AEU - International Journal of Electronics and Communications, 2011, 65 (1): 75- 86.
doi: 10.1016/j.aeue.2010.02.008
22 UHLIG S , QUOITIN B , LEPROPRE J , et al. Providing public intradomain traffic matrices to the research community[J]. Acm Sigcomm Computer Communication Review, 2006, 36 (1): 83- 86.
doi: 10.1145/1111322
23 KINGMA D P, BA J. Adam: a method for stochastic optimization[EB/OL]. (2017-01-30)[2018-04-15]. https://arxiv.org/abs/1412.6980.
[1] 杨亚楠,夏斌,谢楠,袁文浩. 基于BP神经网络和多元Taylor级数的混合定位算法[J]. 山东大学学报 (工学版), 2019, 49(1): 36-40.
[2] 张希华,卢姗姗,苏建军. 全球能源互联网关键技术专利发展现状与对策[J]. 山东大学学报(工学版), 2017, 47(6): 143-150.
[3] 何其佳,刘振丙,徐涛,蒋淑洁. 基于LBP和极限学习机的脑部MR图像分类[J]. 山东大学学报(工学版), 2017, 47(2): 86-93.
[4] 马文静, 吴东亚, 汤凯, 王东柱. 一种应用于交通领域的对象解析系统设计方法[J]. 山东大学学报(工学版), 2015, 45(4): 10-18.
[5] 王启明, 李战国, 樊爱宛. 基于博弈论的量子蚁群算法[J]. 山东大学学报(工学版), 2015, 45(2): 33-36.
[6] 韩忠明, 吴杨, 谭旭升, 刘雯, 杨伟杰. 社会网络结构洞节点度量指标比较与分析[J]. 山东大学学报(工学版), 2015, 45(1): 1-8.
[7] 孙香花. 基于距离向量的改进WSN路由算法[J]. 山东大学学报(工学版), 2012, 42(6): 25-30.
[8] 刘琪,刘沂训,秦丰林. P2P流媒体系统搭便车行为建模研究[J]. 山东大学学报(工学版), 2012, 42(6): 31-36.
[9] 季涛,李永忠. 基于可信计算机制的云计算盲数据处理[J]. 山东大学学报(工学版), 2012, 42(5): 30-34.
[10] 刘东慧1,2,姜薇1*. 基于事件本体的Web不良信息挖掘[J]. 山东大学学报(工学版), 2012, 42(5): 35-40.
[11] 丁彦,李永忠*. 基于PCA和半监督聚类的入侵检测算法研究[J]. 山东大学学报(工学版), 2012, 42(5): 41-46.
[12] 蔡晓军1 ,张擎1 ,柴乔林1 ,孔苏丽2 . 基于能量均衡的n分多路径路由算法[J]. 山东大学学报(工学版), 2009, 39(2): 141-145.
[13] 陈冬岩. 基于多信道的MAC层协议在无线传感器网络中的应用[J]. 山东大学学报(工学版), 2009, 39(1): 41-49.
[14] 蔡忠欣,张华忠 . 一种基于睡眠和网关选择机制的分簇协议[J]. 山东大学学报(工学版), 2008, 38(1): 56-60 .
[15] 黄忠, 葛连升. 基于CoAP的物联网Web服务统一访问方法[J]. 山东大学学报(工学版), 2014, 44(4): 16-21.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 王素玉,艾兴,赵军,李作丽,刘增文 . 高速立铣3Cr2Mo模具钢切削力建模及预测[J]. 山东大学学报(工学版), 2006, 36(1): 1 -5 .
[2] 张永花,王安玲,刘福平 . 低频非均匀电磁波在导电界面的反射相角[J]. 山东大学学报(工学版), 2006, 36(2): 22 -25 .
[3] 李 侃 . 嵌入式相贯线焊接控制系统开发与实现[J]. 山东大学学报(工学版), 2008, 38(4): 37 -41 .
[4] 李梁,罗奇鸣,陈恩红. 对象级搜索中基于图的对象排序模型(英文)[J]. 山东大学学报(工学版), 2009, 39(1): 15 -21 .
[5] 陈瑞,李红伟,田靖. 磁极数对径向磁轴承承载力的影响[J]. 山东大学学报(工学版), 2018, 48(2): 81 -85 .
[6] 王波,王宁生 . 机电装配体拆卸序列的自动生成及组合优化[J]. 山东大学学报(工学版), 2006, 36(2): 52 -57 .
[7] 季涛,高旭,孙同景,薛永端,徐丙垠 . 铁路10 kV自闭/贯通线路故障行波特征分析[J]. 山东大学学报(工学版), 2006, 36(2): 111 -116 .
[8] 浦剑1 ,张军平1 ,黄华2 . 超分辨率算法研究综述[J]. 山东大学学报(工学版), 2009, 39(1): 27 -32 .
[9] 秦通,孙丰荣*,王丽梅,王庆浩,李新彩. 基于极大圆盘引导的形状插值实现三维表面重建[J]. 山东大学学报(工学版), 2010, 40(3): 1 -5 .
[10] 刘文亮,朱维红,陈涤,张泓泉. 基于雷达图像的运动目标形态检测及跟踪技术[J]. 山东大学学报(工学版), 2010, 40(3): 31 -36 .