MACHINE LEARNING AT THE EDGE: A DATA-DRIVEN ARCHITECTURE WITH APPLICATIONS TO 5G CELLULAR NETWORKS

AUTHOR(S)

S.Maheswari, B.Deepa, S.Keetha, G.Suganthi

DOI: https://doi.org/10.46647/ijetms.2023.v07i02.075

ABSTRACT
To meet the ultra-low latency demands of future applications, the fifth generation of cellular networks (5G) will rely on edge cloud installations. In this research, we show that similar deployments may also be deployed in mobile networks to allow sophisticated data-driven and Machine Learning (ML) applications. We propose an edge- controller-based cellular network design and assess its performance using real-world data from hundreds of base stations of a large US operator. In this context, we will discuss how to dynamically cluster and associate base stations and controllers based on users' worldwide movement patterns. The controllers will then be used to run ML algorithms to forecast the number of users in each base station, as well as a use case in which these predictions are utilized by a higher-layer application to direct vehicular traffic based on network Key Performance Indicators (KPIs). We demonstrate that prediction accuracy increases when based on machine learning algorithms that depend on the controllers' view and, as a result, on the spatial correlation provided by user mobility, compared to when the prediction is based only on the local data of each individual base station.

Page No: 668 - 686

References:

[1] "NGMN perspective on 5G architecture," IEEE 81st Veh, by M. Iwamura.Technol.Conf.,2015, pp.1-5.
[2] "Five Disruptive Technology Directions for 5G," by F. Boccardi, R. W. Heath, A. Lozano, T.
L. Marzetta, and P. Popovski, IEEE Communications Magazine, vol.52, no.2, feb. 2014, pages74-80.
[3] "Millimeter-wave cellular wireless networks:" by S. Rangan, T. S. Rappaport, and E. Erkip.Potentials and Obstacles," ProceedingsVol. IEEE102, no.3, pages, March 2014366-385.
[4] "Cognition-based networks:" by M. Zorzi, A. Zanella, A. Testolin, M. D. F. D. Grazia, and
M. Zorzi,A fresh perspective on network optimization that incorporates distributed intelligence and learning," IEEE Access, vol.3, pp.1512-1530, 2015.
[5] R. Li and others, "5G Intelligent:"When Artificial Intelligence Meets Cellular Networks," IEEE Wireless Communications, vol.24, no.5, Oct. 2017, pp.175-183.
[6] S. Chinchali and colleagues,Utilizing deep reinforcement learning to schedule traffic on the cellular network," published in the Proceedings of the National Conference on Artificial Intelligence in 2018, pp.766-774.
[7] "Big data analytics in mobile cellular networks," by Y. He, F. R. Yu, N. Zhao, H. Yin, H. Yao, and R. C. Qiu, IEEE Access, vol.4, no.4, 2016, pp.1985-1996.
[8] A survey of anticipatory mobile networking: N. Bui, M. Cesana, S. A. Hussein, Q. Liao, I. Balanchine, and J. Widmer,"Context-based classification, prediction approaches, and optimization strategies," IEEE Communications Society Transactions, vol.19, no.3, pp.1790- 1821, 2017 Third Quarter
[9] "Anticipatory mobile computing:" by V. Pejovic and M. Musolesi,A review of current research and challenges," ACM ComputingSurv.,vol.47, no.3, pages, April 201547:1-47:29.
[10] "Machine learning paradigms for next-generation wireless networks," IEEE Wireless Communications, C. Jiang, H. Zhang, Y. Ren, Z. Han, K. C. Chen, and L. Hanzo,vol.24, no.2, April 2017.
[11] "5G Challenges:" by A. Imran, A. Zoha, and A. Abu-Dayya."How to Enable 5G by Empowering SON with Big Data," IEEE Network, vol.28, no.6, pp.27-33, Nov. 2014.
[12] "TCP Ex Machine:" by K. Winstein and H. Balakrishnan.In the Proceedings of the ACM SIGCOMM Conf., "Computer generated Congestion Control,"2013, pp.123-134.
[13] "D-DASH:" by M. Gadaleta, A. Zanella, F. Chiariotti, M. Rossi, and othersA deep Q- Learning framework for streaming DASH video," IEEE Transactions on Cognitive Communications Networks, vol.3, no.4, pp.703-718, Dec. 2017.
[14] R. Becker and others,Characterization of human movement using data from cellular networks," ACM Communications, vol.56, no.January 1, 2013, pages74-82.
[15] "Modeling cellular user movement using a leap graph," by W. Dong, N. Duffield, Z. Ge, S. Lee, and J. Pang, in Proc.Int.Conf.Measurement of Passive Active Networks, 2013, pp.53-62.
[16] T. X. Tran, A. Hajisami, P. Pandey, and D. Pompili, "Cooperative versatile edge figuring in 5G organizations:"New paradigms, situations, and difficulties," IEEE Communications Magazine, vol.55, no.April 4, 2017, pages54-61.
[17] "Living on the Edge:" by E. Bastug, M. Bennis, and M. Debbah.The Function of Proactive Caching in 5G Wireless Networks," published in IEEE Communications Magazine, vol.52, no.8, August 2014, pp.82-89.
[18] "Rededge:" by M. Medhat Gaber, P. P. Jayaraman, S. U. R. Malik, A. U. R. Khan, and M. Habib urRehman.A Novel Architecture for the Processing of Big Data in Mobile Edge Computing Environments," vol.6, no.3, 2017.
[19] "Fog-computing-based radio access networks:" by M. Peng, S. Yan, K. Zhang, and C. Wang.issues and difficulties," IEEE Network, vol.30, no.4, pages, July 2016,46-53.
[20] "Learning approaches for long-term channel gain prediction in wireless networks," authored by F. Chiariotti, D. D. Testa, M. Polese, A. Zanella, G. M. D. Nunzio, and M. Zorzi, and published in Proc.Int.Conf.Comput.

How to Cite This Article:
S.Maheswari, B.Deepa, S.Keetha, G.Suganthi . MACHINE LEARNING AT THE EDGE: A DATA-DRIVEN ARCHITECTURE WITH APPLICATIONS TO 5G CELLULAR NETWORKS . ijetms;7(2):668-686. DOI: 10.46647/ijetms.2023.v07i02.075