Traffic Classification using Deep Learning Approach for End-to-End Slice Management in 5G/B5G

Noor Mohammedali; Triantafyllos Kanakis; Ali Al-Sherbaz; Michael Opoku Agyeman

doi:10.1109/ICTC55196.2022.9952446

Traffic Classification using Deep Learning Approach for End-to-End Slice Management in 5G/B5G

Noor Mohammedali^*, Triantafyllos Kanakis, Ali Al-Sherbaz, Michael Opoku Agyeman

^*Corresponding author for this work

Research output: Contribution to Book/Report › Conference Contribution › peer-review

Abstract

Network slicing is a key role in future networks. 5G networks are intended to meet the different service demands of an application offered to users. The 5G architecture is used to match the requirement of the Quality of Service (QoS) by addressing different scenarios in terms of latency, scalability and throughput with different service types. Using machine learning with network slicing allows network operators to create multiple virtual networks or slices on the same physical infrastructure. These slices are independent and customized. Precisely, it will manage dynamically according to the requirements defined between the network operators and the users. For this research, multi-machine learning algorithms are used to train our model, classify the traffic and predict accurate slice type for each user. After the traffic classification, we compared and analysed the performance of various machine learning algorithms in terms of learning percentage, accuracy, precision and F1 score.

Original language	English
Title of host publication	2022 International Conference on Information and Communication Technology Convergence (ICTC)
Publisher	IEEE
Pages	357-362
Number of pages	6
ISBN (Electronic)	978-1-6654-9939-2
DOIs	https://doi.org/10.1109/ICTC55196.2022.9952446
Publication status	Published - 25 Nov 2022
Event	2022 13th International Conference on Information and Communication Technology Convergence - Ramada Plaza Hotel, Jeju Island, Korea, Democratic People's Republic of Duration: 19 Oct 2022 → 21 Oct 2022 https://ictc.org/

Publication series

Name	2022 13th International Conference on Information and Communication Technology Convergence (ICTC)
Publisher	IEEE

Conference

Conference	2022 13th International Conference on Information and Communication Technology Convergence
Abbreviated title	ICTC2022
Country/Territory	Korea, Democratic People's Republic of
City	Jeju Island
Period	19/10/22 → 21/10/22
Internet address	https://ictc.org/

Keywords

5G
End-to-End
Network Slicing
NSSF
Machine Learning
Network Services

Access to Document

10.1109/ICTC55196.2022.9952446

Mohammedali_etal_IEEE_2022_Traffic_Classification_using_Deep_Learning_Approach_for_End-to-End_Slice_ManagementAccepted author manuscript, 798 KB

https://ieeexplore.ieee.org/document/9952446

Cite this

Mohammedali, N., Kanakis, T., Al-Sherbaz, A., & Opoku Agyeman, M. (2022). Traffic Classification using Deep Learning Approach for End-to-End Slice Management in 5G/B5G. In 2022 International Conference on Information and Communication Technology Convergence (ICTC) (pp. 357-362). (2022 13th International Conference on Information and Communication Technology Convergence (ICTC)). IEEE. https://doi.org/10.1109/ICTC55196.2022.9952446

Mohammedali, Noor ; Kanakis, Triantafyllos ; Al-Sherbaz, Ali et al. / Traffic Classification using Deep Learning Approach for End-to-End Slice Management in 5G/B5G. 2022 International Conference on Information and Communication Technology Convergence (ICTC). IEEE, 2022. pp. 357-362 (2022 13th International Conference on Information and Communication Technology Convergence (ICTC)).

@inproceedings{aa3f213197bb4a91915f307fbd72eaf4,

title = "Traffic Classification using Deep Learning Approach for End-to-End Slice Management in 5G/B5G",

abstract = "Network slicing is a key role in future networks. 5G networks are intended to meet the different service demands of an application offered to users. The 5G architecture is used to match the requirement of the Quality of Service (QoS) by addressing different scenarios in terms of latency, scalability and throughput with different service types. Using machine learning with network slicing allows network operators to create multiple virtual networks or slices on the same physical infrastructure. These slices are independent and customized. Precisely, it will manage dynamically according to the requirements defined between the network operators and the users. For this research, multi-machine learning algorithms are used to train our model, classify the traffic and predict accurate slice type for each user. After the traffic classification, we compared and analysed the performance of various machine learning algorithms in terms of learning percentage, accuracy, precision and F1 score.",

keywords = "5G, End-to-End, Network Slicing, NSSF, Machine Learning, Network Services",

author = "Noor Mohammedali and Triantafyllos Kanakis and Ali Al-Sherbaz and {Opoku Agyeman}, Michael",

year = "2022",

month = nov,

day = "25",

doi = "10.1109/ICTC55196.2022.9952446",

language = "English",

series = "2022 13th International Conference on Information and Communication Technology Convergence (ICTC)",

publisher = "IEEE",

pages = "357--362",

booktitle = "2022 International Conference on Information and Communication Technology Convergence (ICTC)",

note = "2022 13th International Conference on Information and Communication Technology Convergence, ICTC2022 ; Conference date: 19-10-2022 Through 21-10-2022",

url = "https://ictc.org/",

}

Mohammedali, N , Kanakis, T, Al-Sherbaz, A & Opoku Agyeman, M 2022, Traffic Classification using Deep Learning Approach for End-to-End Slice Management in 5G/B5G. in 2022 International Conference on Information and Communication Technology Convergence (ICTC). 2022 13th International Conference on Information and Communication Technology Convergence (ICTC), IEEE, pp. 357-362, 2022 13th International Conference on Information and Communication Technology Convergence, Jeju Island, Korea, Democratic People's Republic of, 19/10/22. https://doi.org/10.1109/ICTC55196.2022.9952446

Traffic Classification using Deep Learning Approach for End-to-End Slice Management in 5G/B5G. / Mohammedali, Noor ; Kanakis, Triantafyllos; Al-Sherbaz, Ali et al.
2022 International Conference on Information and Communication Technology Convergence (ICTC). IEEE, 2022. p. 357-362 (2022 13th International Conference on Information and Communication Technology Convergence (ICTC)).

Research output: Contribution to Book/Report › Conference Contribution › peer-review

TY - GEN

T1 - Traffic Classification using Deep Learning Approach for End-to-End Slice Management in 5G/B5G

AU - Mohammedali, Noor

AU - Kanakis, Triantafyllos

AU - Al-Sherbaz, Ali

AU - Opoku Agyeman, Michael

PY - 2022/11/25

Y1 - 2022/11/25

N2 - Network slicing is a key role in future networks. 5G networks are intended to meet the different service demands of an application offered to users. The 5G architecture is used to match the requirement of the Quality of Service (QoS) by addressing different scenarios in terms of latency, scalability and throughput with different service types. Using machine learning with network slicing allows network operators to create multiple virtual networks or slices on the same physical infrastructure. These slices are independent and customized. Precisely, it will manage dynamically according to the requirements defined between the network operators and the users. For this research, multi-machine learning algorithms are used to train our model, classify the traffic and predict accurate slice type for each user. After the traffic classification, we compared and analysed the performance of various machine learning algorithms in terms of learning percentage, accuracy, precision and F1 score.

AB - Network slicing is a key role in future networks. 5G networks are intended to meet the different service demands of an application offered to users. The 5G architecture is used to match the requirement of the Quality of Service (QoS) by addressing different scenarios in terms of latency, scalability and throughput with different service types. Using machine learning with network slicing allows network operators to create multiple virtual networks or slices on the same physical infrastructure. These slices are independent and customized. Precisely, it will manage dynamically according to the requirements defined between the network operators and the users. For this research, multi-machine learning algorithms are used to train our model, classify the traffic and predict accurate slice type for each user. After the traffic classification, we compared and analysed the performance of various machine learning algorithms in terms of learning percentage, accuracy, precision and F1 score.

KW - 5G

KW - End-to-End

KW - Network Slicing

KW - NSSF

KW - Machine Learning

KW - Network Services

U2 - 10.1109/ICTC55196.2022.9952446

DO - 10.1109/ICTC55196.2022.9952446

M3 - Conference Contribution

T3 - 2022 13th International Conference on Information and Communication Technology Convergence (ICTC)

SP - 357

EP - 362

BT - 2022 International Conference on Information and Communication Technology Convergence (ICTC)

PB - IEEE

T2 - 2022 13th International Conference on Information and Communication Technology Convergence

Y2 - 19 October 2022 through 21 October 2022

ER -

Mohammedali N , Kanakis T, Al-Sherbaz A, Opoku Agyeman M. Traffic Classification using Deep Learning Approach for End-to-End Slice Management in 5G/B5G. In 2022 International Conference on Information and Communication Technology Convergence (ICTC). IEEE. 2022. p. 357-362. (2022 13th International Conference on Information and Communication Technology Convergence (ICTC)). Epub 2022 Nov 25. doi: 10.1109/ICTC55196.2022.9952446

Traffic Classification using Deep Learning Approach for End-to-End Slice Management in 5G/B5G

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this