A Unified Framework for 5G Network Management Tools

Jukka K. Nurminen, Harrison Mfula

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

Abstract

Limitations in the software architecture of current network management tools such as lack of support for combined batch and real time data processing, adaptive machine learning, support for heterogeneous data sources and the fragmentation of emerging solutions needs to be addressed in order to create a solid and forward leaning foundation for implementing 5G solutions. To address these limitations, this paper introduces the extended lambda architecture (ELA). It focuses on bringing agility and continuous learning based decision making support into the design of a unified architectural framework for new network management tools by combining batch and real time data processing with adaptive machine learning in a simple Monitor-Analyze-Plan-Execute over a shared Knowledge (MAPE-K) loop. The benefits of using this architecture are evaluated using a proof of concept (PoC) implementation of a reliable and proactive tool for detection and compensation of cell outages in a simulated 5G network.

Original languageEnglish
Title of host publicationProceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages41-48
Number of pages8
ISBN (Electronic)978-1-5386-9133-5
DOIs
Publication statusPublished - 2 Jan 2019
MoE publication typeA4 Article in a conference publication
Event11th IEEE International Conference on Service-Oriented Computing and Applications, SOCA 2018 - Paris, France
Duration: 20 Nov 201822 Nov 2018

Publication series

NameProceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018

Conference

Conference11th IEEE International Conference on Service-Oriented Computing and Applications, SOCA 2018
CountryFrance
CityParis
Period20/11/1822/11/18

Fingerprint

Network management
Learning systems
Software architecture
Outages
Decision making
Management tools
Real-time data
Batch
Machine learning
Data sources
Fragmentation
Shared knowledge
Agility

Keywords

  • 5G
  • Big data
  • Lambda architecture
  • Machine learning
  • Self-organizing networks

Cite this

Nurminen, J. K., & Mfula, H. (2019). A Unified Framework for 5G Network Management Tools. In Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018 (pp. 41-48). [8599577] Institute of Electrical and Electronics Engineers Inc.. Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018 https://doi.org/10.1109/SOCA.2018.00013
Nurminen, Jukka K. ; Mfula, Harrison. / A Unified Framework for 5G Network Management Tools. Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 41-48 (Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018).
@inproceedings{02b382e19b17446b9c7e8783a0d92a9d,
title = "A Unified Framework for 5G Network Management Tools",
abstract = "Limitations in the software architecture of current network management tools such as lack of support for combined batch and real time data processing, adaptive machine learning, support for heterogeneous data sources and the fragmentation of emerging solutions needs to be addressed in order to create a solid and forward leaning foundation for implementing 5G solutions. To address these limitations, this paper introduces the extended lambda architecture (ELA). It focuses on bringing agility and continuous learning based decision making support into the design of a unified architectural framework for new network management tools by combining batch and real time data processing with adaptive machine learning in a simple Monitor-Analyze-Plan-Execute over a shared Knowledge (MAPE-K) loop. The benefits of using this architecture are evaluated using a proof of concept (PoC) implementation of a reliable and proactive tool for detection and compensation of cell outages in a simulated 5G network.",
keywords = "5G, Big data, Lambda architecture, Machine learning, Self-organizing networks",
author = "Nurminen, {Jukka K.} and Harrison Mfula",
year = "2019",
month = "1",
day = "2",
doi = "10.1109/SOCA.2018.00013",
language = "English",
series = "Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "41--48",
booktitle = "Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018",
address = "United States",

}

Nurminen, JK & Mfula, H 2019, A Unified Framework for 5G Network Management Tools. in Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018., 8599577, Institute of Electrical and Electronics Engineers Inc., Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018, pp. 41-48, 11th IEEE International Conference on Service-Oriented Computing and Applications, SOCA 2018, Paris, France, 20/11/18. https://doi.org/10.1109/SOCA.2018.00013

A Unified Framework for 5G Network Management Tools. / Nurminen, Jukka K.; Mfula, Harrison.

Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018. Institute of Electrical and Electronics Engineers Inc., 2019. p. 41-48 8599577 (Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018).

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - A Unified Framework for 5G Network Management Tools

AU - Nurminen, Jukka K.

AU - Mfula, Harrison

PY - 2019/1/2

Y1 - 2019/1/2

N2 - Limitations in the software architecture of current network management tools such as lack of support for combined batch and real time data processing, adaptive machine learning, support for heterogeneous data sources and the fragmentation of emerging solutions needs to be addressed in order to create a solid and forward leaning foundation for implementing 5G solutions. To address these limitations, this paper introduces the extended lambda architecture (ELA). It focuses on bringing agility and continuous learning based decision making support into the design of a unified architectural framework for new network management tools by combining batch and real time data processing with adaptive machine learning in a simple Monitor-Analyze-Plan-Execute over a shared Knowledge (MAPE-K) loop. The benefits of using this architecture are evaluated using a proof of concept (PoC) implementation of a reliable and proactive tool for detection and compensation of cell outages in a simulated 5G network.

AB - Limitations in the software architecture of current network management tools such as lack of support for combined batch and real time data processing, adaptive machine learning, support for heterogeneous data sources and the fragmentation of emerging solutions needs to be addressed in order to create a solid and forward leaning foundation for implementing 5G solutions. To address these limitations, this paper introduces the extended lambda architecture (ELA). It focuses on bringing agility and continuous learning based decision making support into the design of a unified architectural framework for new network management tools by combining batch and real time data processing with adaptive machine learning in a simple Monitor-Analyze-Plan-Execute over a shared Knowledge (MAPE-K) loop. The benefits of using this architecture are evaluated using a proof of concept (PoC) implementation of a reliable and proactive tool for detection and compensation of cell outages in a simulated 5G network.

KW - 5G

KW - Big data

KW - Lambda architecture

KW - Machine learning

KW - Self-organizing networks

UR - http://www.scopus.com/inward/record.url?scp=85061529786&partnerID=8YFLogxK

U2 - 10.1109/SOCA.2018.00013

DO - 10.1109/SOCA.2018.00013

M3 - Conference article in proceedings

T3 - Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018

SP - 41

EP - 48

BT - Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Nurminen JK, Mfula H. A Unified Framework for 5G Network Management Tools. In Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018. Institute of Electrical and Electronics Engineers Inc. 2019. p. 41-48. 8599577. (Proceedings - IEEE 11th International Conference on Service-Oriented Computing and Applications, SOCA 2018). https://doi.org/10.1109/SOCA.2018.00013