Applying Machine Learning to LTE Traffic Prediction: Comparison of Bagging, Random Forest, and SVM

Nikolai Stepanov; Daria Alekseeva; Aleksandr Ometov; Elena-Simona Lohan

doi:10.1109/icumt51630.2020.9222418

Applying Machine Learning to LTE Traffic Prediction: Comparison of Bagging, Random Forest, and SVM

Nikolai Stepanov, Daria Alekseeva, Aleksandr Ometov, Elena-Simona Lohan

Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

35 Citations (Scopus)

217 Downloads (Pure)

Abstract

Today, a significant share of smartphone applications use Artificial Intelligence (AI) elements that, in turn, are based on Machine Learning (ML) principles. Particularly, ML is also applied to the Edge paradigm aiming to predict and optimize the network load conventionally caused by human-based traffic, which is growing each year rapidly. The application of both standard and deep ML techniques is expected to improve the networks’ operation in the most complex heterogeneous environment. In this work, we propose a method to predict the LTE network edge traffic by utilizing various ML techniques. The analysis is based on the public cellular traffic dataset, and it presents a comparison of the quality metrics. The Support Vector Machines method allows much faster training than the Bagging and Random Forest that operate well with a mixture of numerical and categorical features.

Original language	English
Title of host publication	12th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)
Publisher	IEEE
Pages	119-123
Number of pages	5
ISBN (Print)	9781728192819
DOIs	https://doi.org/10.1109/icumt51630.2020.9222418
Publication status	Published - 14 Oct 2020
Publication type	A4 Article in conference proceedings
Event	International Congress on Ultra Modern Telecommunications and Control Systems and Workshops - Brno, Czech Republic Duration: 5 Oct 2020 → 7 Oct 2020

Publication series

Name	International Conference on Ultra Modern Telecommunications & workshops
ISSN (Electronic)	2157-023X

Conference

Conference	International Congress on Ultra Modern Telecommunications and Control Systems and Workshops
Country/Territory	Czech Republic
City	Brno
Period	5/10/20 → 7/10/20

Keywords

Machine Learning
traffic analysis
LTE
NGN
optimisation

Publication forum classification

Publication forum level 1

ASJC Scopus subject areas

Computer Networks and Communications
Control and Systems Engineering

Access to Document

10.1109/icumt51630.2020.9222418

Applying_Machine_Learning_to_LTE_Traffic_Prediction_Comparison_of_Bagging_Random_Forest_and_SVM
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Accepted author manuscript, 238 KB

http://urn.fi/URN:NBN:fi:tuni-202101121182

Cite this

Stepanov, N., Alekseeva, D., Ometov, A., & Lohan, E.-S. (2020). Applying Machine Learning to LTE Traffic Prediction: Comparison of Bagging, Random Forest, and SVM. In 12th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT) (pp. 119-123). (International Conference on Ultra Modern Telecommunications & workshops). IEEE. https://doi.org/10.1109/icumt51630.2020.9222418

Stepanov, Nikolai ; Alekseeva, Daria ; Ometov, Aleksandr et al. / Applying Machine Learning to LTE Traffic Prediction: Comparison of Bagging, Random Forest, and SVM. 12th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT). IEEE, 2020. pp. 119-123 (International Conference on Ultra Modern Telecommunications & workshops).

@inproceedings{dfca63f6432f40978602d637042f6884,

title = "Applying Machine Learning to LTE Traffic Prediction: Comparison of Bagging, Random Forest, and SVM",

abstract = "Today, a significant share of smartphone applications use Artificial Intelligence (AI) elements that, in turn, are based on Machine Learning (ML) principles. Particularly, ML is also applied to the Edge paradigm aiming to predict and optimize the network load conventionally caused by human-based traffic, which is growing each year rapidly. The application of both standard and deep ML techniques is expected to improve the networks{\textquoteright} operation in the most complex heterogeneous environment. In this work, we propose a method to predict the LTE network edge traffic by utilizing various ML techniques. The analysis is based on the public cellular traffic dataset, and it presents a comparison of the quality metrics. The Support Vector Machines method allows much faster training than the Bagging and Random Forest that operate well with a mixture of numerical and categorical features.",

keywords = "Machine Learning, traffic analysis, LTE, NGN, optimisation",

author = "Nikolai Stepanov and Daria Alekseeva and Aleksandr Ometov and Elena-Simona Lohan",

note = "jufoid=72315 Funding Information: ACKNOWLEDGMENT This paper is supported by the Doctoral training network in ELectronics, Telecommunications and Automation (DELTA), and by the Academy of Finland ULTRA project (328226/328214). Publisher Copyright: {\textcopyright} 2020 IEEE. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.; International Congress on Ultra Modern Telecommunications and Control Systems and Workshops ; Conference date: 05-10-2020 Through 07-10-2020",

year = "2020",

month = oct,

day = "14",

doi = "10.1109/icumt51630.2020.9222418",

language = "English",

isbn = "9781728192819",

series = "International Conference on Ultra Modern Telecommunications & workshops",

publisher = "IEEE",

pages = "119--123",

booktitle = "12th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)",

address = "United States",

}

Stepanov, N, Alekseeva, D, Ometov, A & Lohan, E-S 2020, Applying Machine Learning to LTE Traffic Prediction: Comparison of Bagging, Random Forest, and SVM. in 12th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT). International Conference on Ultra Modern Telecommunications & workshops, IEEE, pp. 119-123, International Congress on Ultra Modern Telecommunications and Control Systems and Workshops, Brno, Czech Republic, 5/10/20. https://doi.org/10.1109/icumt51630.2020.9222418

Applying Machine Learning to LTE Traffic Prediction: Comparison of Bagging, Random Forest, and SVM. / Stepanov, Nikolai; Alekseeva, Daria; Ometov, Aleksandr et al.
12th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT). IEEE, 2020. p. 119-123 (International Conference on Ultra Modern Telecommunications & workshops).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Applying Machine Learning to LTE Traffic Prediction: Comparison of Bagging, Random Forest, and SVM

AU - Stepanov, Nikolai

AU - Alekseeva, Daria

AU - Ometov, Aleksandr

AU - Lohan, Elena-Simona

N1 - jufoid=72315 Funding Information: ACKNOWLEDGMENT This paper is supported by the Doctoral training network in ELectronics, Telecommunications and Automation (DELTA), and by the Academy of Finland ULTRA project (328226/328214). Publisher Copyright: © 2020 IEEE. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.

PY - 2020/10/14

Y1 - 2020/10/14

N2 - Today, a significant share of smartphone applications use Artificial Intelligence (AI) elements that, in turn, are based on Machine Learning (ML) principles. Particularly, ML is also applied to the Edge paradigm aiming to predict and optimize the network load conventionally caused by human-based traffic, which is growing each year rapidly. The application of both standard and deep ML techniques is expected to improve the networks’ operation in the most complex heterogeneous environment. In this work, we propose a method to predict the LTE network edge traffic by utilizing various ML techniques. The analysis is based on the public cellular traffic dataset, and it presents a comparison of the quality metrics. The Support Vector Machines method allows much faster training than the Bagging and Random Forest that operate well with a mixture of numerical and categorical features.

AB - Today, a significant share of smartphone applications use Artificial Intelligence (AI) elements that, in turn, are based on Machine Learning (ML) principles. Particularly, ML is also applied to the Edge paradigm aiming to predict and optimize the network load conventionally caused by human-based traffic, which is growing each year rapidly. The application of both standard and deep ML techniques is expected to improve the networks’ operation in the most complex heterogeneous environment. In this work, we propose a method to predict the LTE network edge traffic by utilizing various ML techniques. The analysis is based on the public cellular traffic dataset, and it presents a comparison of the quality metrics. The Support Vector Machines method allows much faster training than the Bagging and Random Forest that operate well with a mixture of numerical and categorical features.

KW - Machine Learning

KW - traffic analysis

KW - LTE

KW - NGN

KW - optimisation

U2 - 10.1109/icumt51630.2020.9222418

DO - 10.1109/icumt51630.2020.9222418

M3 - Conference contribution

SN - 9781728192819

T3 - International Conference on Ultra Modern Telecommunications & workshops

SP - 119

EP - 123

BT - 12th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)

PB - IEEE

T2 - International Congress on Ultra Modern Telecommunications and Control Systems and Workshops

Y2 - 5 October 2020 through 7 October 2020

ER -

Stepanov N, Alekseeva D, Ometov A , Lohan ES. Applying Machine Learning to LTE Traffic Prediction: Comparison of Bagging, Random Forest, and SVM. In 12th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT). IEEE. 2020. p. 119-123. (International Conference on Ultra Modern Telecommunications & workshops). doi: 10.1109/icumt51630.2020.9222418

Applying Machine Learning to LTE Traffic Prediction: Comparison of Bagging, Random Forest, and SVM

Abstract

Publication series

Conference

Keywords

Publication forum classification

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this