计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (6A): 241100058-7.doi: 10.11896/jsjkx.241100058

• 大语言模型技术及应用 • 上一篇    下一篇

基于大语言模型的网络流量智能预测

周磊1,2, 石怀峰1,2,3, 杨恺1,2, 王睿2,4, 刘超凡1,2   

  1. 1 南京信息工程大学电子与信息工程学院 南京 210044
    2 南京信息工程大学复杂环境智能保障技术教育部重点实验室 南京 210044
    3 国防科技大学第六十三研究所 南京 210007
    4 南京信息工程大学计算机学院、网络空间安全学院 南京 210044
  • 出版日期:2025-06-16 发布日期:2025-06-12
  • 通讯作者: 石怀峰(shihuaifeng@nuist.edu.cn)
  • 作者简介:(202283830003@nuist.edu.cn)
  • 基金资助:
    南京信息工程大学复杂环境智能保障技术教育部重点实验室开放基金(B2202401);南京信息工程大学人才启动经费(1083142401004)

Intelligent Prediction of Network Traffic Based on Large Language Model

ZHOU Lei1,2, SHI Huaifeng1,2,3, YANG Kai1,2, WANG Rui2,4, LIU Chaofan1,2   

  1. 1 School of Electronics and Information Engineering,Nanjing University of Information Science and Technology,Nanjing 210044,China
    2 Key Laboratory of Intelligent Support Technology for Complex Environments,Ministry of Education,Nanjing University of Information Science and Technology,Nanjing 210044,China
    3 63rd Research Institute of National University of Defense Technology,Nanjing 210007,China
    4 School of Computer Science,School of Cyberspace Science and Engineering,Nanjing University of Information Science and Technology,Nanjing 210044,China
  • Online:2025-06-16 Published:2025-06-12
  • About author:ZHOU Lei,born in 2003.His main research interests include network traffic prediction technology and so on.
    SHI Huaifeng,born in 1989,Ph.D,associate professor,is amember of CCF(No.39243M).His main research interests include the theory and key technologies of space-ground integrated intelligent networks and so on.
  • Supported by:
    Key Laboratory of Intelligent Support Technology for Complex Environments,Ministry of Education(B2202401) and Startup Foundation for Introducing Talent,Nanjing University of Information Science and Technology(1083142401004).

PDF (PC)

57

摘要： 随着5G基站数量的倍增和接入终端数量的剧增,网络流量的规模将呈现指数级增长,网络流量则呈现出显著的非线性、多模态和突发性特征,对网络资源分配和优化提出了新的挑战。为应对这些挑战,提出了一种基于大语言模型(LLM)的网络流量预测方法(NT-LLM)。该方法通过重编程技术,将传统的网络流量数据转换为适合LLM处理的形式,从而充分利用LLM在跨任务推理和复杂模式识别方面的优势,仅需少量训练数据和较短训练周期,就能够高效处理不同时间尺度的复杂网络流量模式。实验结果表明,与LSTM,Informer,Transformer等基线模型相比,NT-LLM模型在多个区域的网络流量预测均方误差显著下降,分别降低了44.26%,56.78%和51.36%。此外,该方法无需对预训练的语言模型进行大规模微调,具有较强的扩展性和适应性,能够在减少计算资源消耗的同时保持高精度的预测能力。

关键词: 网络流量预测, 大语言模型, 重编程, 时间序列数据, 深度学习

Abstract: With the exponential growth in the number of 5G base stations and the surge in connected terminals,the scale of network traffic is expected to grow exponentially,exhibiting significant nonlinear,multimodal,and bursty characteristics,posing new challenges to network resource allocation and optimization..To address these challenges,this paper proposed a network traffic prediction method based on large language models(NT-LLM).This approach leverages reprogramming techniques to transform traditional network traffic data into a format suitable for LLMs,thus fully utilizing their advantages in cross-task reasoning and complex pattern recognition.With only a small amount of training data and a short training period,NT-LLM can efficiently handle complex network traffic patterns at different time scales.Experimental results demonstrate that compared to baseline models such as LSTM,Informer,and Transformer,the NT-LLM model significantly reduces the mean squared error of network traffic predictions across multiple regions by 44.26%,56.78%,and 51.36%,respectively.Furthermore,this method does not require extensive fine-tuning of pre-trained language models,showcasing strong scalability and adaptability.It maintains high prediction accuracy while reducing computational resource consumption.

Key words: Network traffic prediction, Large language model, Reprogramming, Time-series data, Deep learning

中图分类号: 

  • TP393
[1]DENG L,RUAN K,CHEN X,et al.An IP network traffic prediction method based on ARIMA and n-BEATS[C]//International Conference on Power,Intelligent Computing and Systems(ICPICS).IEEE,2022:336-341.
[2]WANG J.A process level network traffic prediction algorithm based on ARIMA model in smart substation[C]//International Conference on Signal Processing,Communication and Computing(ICSPCC 2013).IEEE,2013:1-5.
[3]ABBASI M,SHAHRAKI A,TAHERKORDI A.Deep learning for network traffic monitoring and analysis(NTMA):A survey[J].Computer Communications,2021,170:19-41.
[4]TAHAEI H,AFIFI F,ASEMI A,et al.The rise of traffic classification in IoT networks:A survey[J].Journal of Network and Computer Applications,2020,154:102538.
[5]DAMASEVICIUS R,VENCKAUSKAS A,GRIGALIUNAS S,et al.LITNET-2020:An annotated real-world network flow dataset for network intrusion detection[J].Electronics,2020,9(5):800.
[6]SATRIO C B A,DARMAWAN W,NADIA B U,et al.Time series analysis and forecasting of coronavirus disease in Indonesia using ARIMA model and PROPHET[J].Procedia Computer Science,2021,179:524-532.
[7]YAN Q,SHENG H,RONG H.Research on Modeling and Forecasting of Network Traffic Data with Alpha Distribution Noise Based on FARIMA Model[C]//IOP Conference Series:Mate-rials Science and Engineering.IOP Publishing,2020,750(1):012149.
[8]MÜNGEN A A,TAŞ İ Ç.An investigation about traffic prediction by using ANN and SVM algorithms[C]//International Conference on Electrical,Communication,and Computer Engineering(ICECCE).IEEE,2021:1-6.
[9]STEPANOV N,ALEKSEEVA D,OMETOV A,et al.Applying machine learning to LTE traffic prediction:Comparison of bagging,random forest,and SVM[C]//International Congress on Ultra Modern Telecommunications and Control Systems and Workshops(ICUMT).IEEE,2020:119-123.
[10]LIU X,FANG X,QIN Z,et al.A short-term forecasting algorithm for network traffic based on chaos theory and SVM[J].Journal of Network and Systems Management,2011,19:427-447.
[11]ULANOWICZ B,DOPART D,KNAPINSKA A,et al.Combining Random Forest and Linear Regression to Improve Network Traffic Prediction[C]//2023 23rd International Conference on Transparent Optical Networks(ICTON).IEEE,2023:1-4.
[12]SEPASGOZZAR S S,PIERRE S.Network traffic predictionmodel considering road traffic parameters using artificial intelligence methods in VANET[J].IEEE Access,2022,10:8227-8242.
[13]STRYCZEK S,NATKANIEC M.Internet threat detection insmart grids based on network traffic analysis using lstm,if,and svm[J].Energies,2022,16(1):329.
[14]BI J,ZHANG X,YUAN H,et al.A hybrid prediction method for realistic network traffic with temporal convolutional network and LSTM[J].IEEE Transactions on Automation Science and Engineering,2021,19(3):1869-1879.
[15]BI J,YUAN H,XU K,et al.Large-scale network traffic prediction with LSTM and temporal convolutional networks[C]//2022 International Conference on Robotics and Automation(ICRA).IEEE,2022:3865-3870.
[16]WAN X,LIU H,XU H,et al.Network traffic prediction based on LSTM and transfer learning[J].IEEE Access,2022,10:86181-86190.
[17]LIU B,TANG X,CHENG J,et al.Traffic flow combinationforecasting method based on improved LSTM and ARIMA[J].International Journal of Embedded Systems,2020,12(1):22-30.
[18]MA X,ZHONG H,LI Y,et al.Forecasting transportation network speed using deep capsule networks with nested LSTM models[J].IEEE Transactions on Intelligent Transportation Systems,2020,22(8):4813-4824.
[19]GUO Y,PENG Y,HAO R,et al.Capturing spatial-temporalcorrelations with Attention based Graph Convolutional Network for network traffic prediction[J].Journal of Network and Computer Applications,2023,220:103746.
[20]BRIMOS P,KARAMANOU A,KALAMPOKIS E,et al.Graph neural networks and open-government data to forecast traffic flow[J].Information,2023,14(4):228.
[21]LI F,FENG J,YAN H,et al.Dynamic graph convolutional recurrent network for traffic prediction:Benchmark and solution[J].ACM Transactions on Knowledge Discovery from Data,2023,17(1):1-21.
[22]XIE Y,JIN C.Evaluations of Multi-Step Traffic Flow Prediction Models Based on Graph Neural Networks[C]//2024 6th International Conference on Communications,Information System and Computer Engineering(CISCE).IEEE,2024:1100-1104.
[23]HU Y,ZHOU Y,SONG J,et al.Citywide mobile traffic forecas-ting using spatial-temporal downsampling transformer neural networks[J].IEEE Transactions on Network and Service Mana-gement,2022,20(1):152-165.
[24]YIN K,YANG Y,YAO C,et al.Long-term prediction of network security situation through the use of the transformer-based model[J].IEEE Access,2022,10:56145-56157.
[25]ZHAO L,YUAN H,XU K,et al.Hybrid network attack prediction with Savitzky-Golay filter-assisted informer[J].Expert Systems with Applications,2024,235:121126.
[26]ABBASI M,SHAHRAKI A,TAHERKODI A.Deep learning for network traffic monitoring and analysis(NTMA):A survey[J].Computer Communications,2021,170:19-41.
[27]ALIZADEH M,BEHESHTI M T H,RAMEZANI A,et al.Network traffic forecasting based on fixed telecommunication data using deep learning[C]//2020 6th Iranian Conference on Signal Processing and Intelligent Systems(ICSPIS).IEEE,2020:1-7.
[28]QIU L,JIN L,CHAI L.Network traffic prediction based onspatio-temporal graph convolutional network[C]//2023 42nd Chinese Control Conference(CCC).IEEE,2023:8426-8431.
[29]LI X,WEI D,MENG L,et al.Prediction of Electricity Network Traffic Based on BP NeuralNetwork-Simulated Annealing Algorithm[C]//2022 4th International Conference on Smart Power &Internet Energy Systems(SPIES).IEEE,2022:1339-1343.
[30]CONG L,SHI B,DI X,et al.Research on Satellite NetworkTraffic Prediction Algorithm Based on Gray Wolf Algorithm Optimizing GRU and Spatiotemporal Analysis[C]//2023 15th International Conference on Communication Software and Networks(ICCSN).IEEE,2023:123-131.
[31]WANG S,YAN C,SHAO Y.Road Traffic Accident Prediction Model Based on J-LSTM＋Attention Mechanism[C]//2023 6th International Conference on Artificial Intelligence and Big Data(ICAIBD).IEEE,2023:635-638.
[32]HE Q,MOAYYEDI A,DAN G,et al.A meta-learning scheme for adaptive short-term network traffic prediction[J].IEEE Journal on Selected Areas in Communications,2020,38(10):2271-2283.
[33]GUAN B,CAO J,WANG X,et al.Integrated method of deeplearning and large language model in speech recognition[C]//2024 IEEE 7th International Conference on Electronic Information and Communication Technology(ICEICT).IEEE,2024:487-490.
[34]GRUVER N,FINZI M,QIU S,et al.Large language models are zero-shot time series forecasters[J].Advances in Neural Information Processing Systems,2024,36.
[35]DHINGRA B,COLE J R,EISENSCHLOS J M,et al.Time-aware language models as temporal knowledge bases[J].Transactions of the Association for Computational Linguistics,2022,10:257-273.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发