A Machine Learning Methodology for Network Anomalies Detection in O-RAN Networks

Atanas Vlahov; Vladimir Poulkov; Pavlos Lazaridis; Zaharias Zaharis

A Machine Learning Methodology for Network Anomalies Detection in O-RAN Networks

Atanas Vlahov, Vladimir Poulkov, Pavlos Lazaridis, Zaharias Zaharis

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

Abstract

Despite the extensive research and various existing systems for detecting and classifying anomalies in operational networks, Internet Service Providers (ISPs) continue to seek efficient methods to handle the increasing number of network traffic anomalies they encounter in their daily operations. This research paper tackles the challenge of automatically detecting network traffic anomalies using Machine Learning (ML) techniques. We introduce a straightforward classification approach based on Deep Neural Networks and assess its accuracy, precision, and recall performance. To evaluate the proposed neural network, we train and test it using data collected from a real LTE deployment of 10 base stations.

Original language	English
Title of host publication	28th European Wireless Conference, EW 2023
Publisher	VDE Verlag GmbH
Pages	167-171
Number of pages	5
ISBN (Electronic)	9783800762262, 9783800762255
Publication status	Published - 2 Oct 2023
Event	28th European Wireless Conference - Rome, Italy Duration: 2 Oct 2023 → 4 Oct 2023 Conference number: 28

Conference

Conference	28th European Wireless Conference
Abbreviated title	EW 2023
Country/Territory	Italy
City	Rome
Period	2/10/23 → 4/10/23

Access to Document

https://www.vde-verlag.de/proceedings-en/566225028.htmlLicence: Unspecified

Cite this

@inproceedings{732a15ea8e1b4e19b91366c67fff6834,

title = "A Machine Learning Methodology for Network Anomalies Detection in O-RAN Networks",

abstract = "Despite the extensive research and various existing systems for detecting and classifying anomalies in operational networks, Internet Service Providers (ISPs) continue to seek efficient methods to handle the increasing number of network traffic anomalies they encounter in their daily operations. This research paper tackles the challenge of automatically detecting network traffic anomalies using Machine Learning (ML) techniques. We introduce a straightforward classification approach based on Deep Neural Networks and assess its accuracy, precision, and recall performance. To evaluate the proposed neural network, we train and test it using data collected from a real LTE deployment of 10 base stations.",

keywords = "Anomaly detection, Machine learning, Non-realtime RIC, O-RAN, rApp",

author = "Atanas Vlahov and Vladimir Poulkov and Pavlos Lazaridis and Zaharias Zaharis",

note = "Funding Information: This research was supported by the European Union through the Horizon 2020 Marie Sk\u0142odowska-Curie Research and Innovation Staff Exchange \u201CResearch Collaboration and Mobility for Beyond 5G Future Wireless Networks (RECOMBINE)\u201D under grant agreement no. 872857. Publisher Copyright: {\textcopyright} VDE VERLAG GMBH - Berlin - Offenbach.; 28th European Wireless Conference, EW 2023 ; Conference date: 02-10-2023 Through 04-10-2023",

year = "2023",

month = oct,

day = "2",

language = "English",

pages = "167--171",

booktitle = "28th European Wireless Conference, EW 2023",

publisher = "VDE Verlag GmbH",

address = "Germany",

}

TY - GEN

T1 - A Machine Learning Methodology for Network Anomalies Detection in O-RAN Networks

AU - Vlahov, Atanas

AU - Poulkov, Vladimir

AU - Lazaridis, Pavlos

AU - Zaharis, Zaharias

N1 - Conference code: 28

PY - 2023/10/2

Y1 - 2023/10/2

N2 - Despite the extensive research and various existing systems for detecting and classifying anomalies in operational networks, Internet Service Providers (ISPs) continue to seek efficient methods to handle the increasing number of network traffic anomalies they encounter in their daily operations. This research paper tackles the challenge of automatically detecting network traffic anomalies using Machine Learning (ML) techniques. We introduce a straightforward classification approach based on Deep Neural Networks and assess its accuracy, precision, and recall performance. To evaluate the proposed neural network, we train and test it using data collected from a real LTE deployment of 10 base stations.

AB - Despite the extensive research and various existing systems for detecting and classifying anomalies in operational networks, Internet Service Providers (ISPs) continue to seek efficient methods to handle the increasing number of network traffic anomalies they encounter in their daily operations. This research paper tackles the challenge of automatically detecting network traffic anomalies using Machine Learning (ML) techniques. We introduce a straightforward classification approach based on Deep Neural Networks and assess its accuracy, precision, and recall performance. To evaluate the proposed neural network, we train and test it using data collected from a real LTE deployment of 10 base stations.

KW - Anomaly detection

KW - Machine learning

KW - Non-realtime RIC

KW - O-RAN

KW - rApp

UR - http://www.scopus.com/inward/record.url?scp=85191231803&partnerID=8YFLogxK

UR - https://www.vde-verlag.de/buecher/566225/european-wireless-2023.html

M3 - Conference contribution

AN - SCOPUS:85191231803

SP - 167

EP - 171

BT - 28th European Wireless Conference, EW 2023

PB - VDE Verlag GmbH

T2 - 28th European Wireless Conference

Y2 - 2 October 2023 through 4 October 2023

ER -

A Machine Learning Methodology for Network Anomalies Detection in O-RAN Networks

Abstract

Conference

Access to Document

Other files and links

Cite this