1. Home
    Item seperator
  3. Item seperator
    AI for Self Organized Networks

We develop learning methods that optimize the performance and energy efficiency of next generation wireless networks. We demonstrate the effectiveness of our approaches via numerous field trials in collaboration with Nokia's Mobile Networks business.

On this page

Beamforming

We optimize 5G downlink and uplink beams to ensure maximum network coverage and optimal user performance. We are capable of predicting user geographical distribution and estimating the evolution of the channel over time on a user basis.

Beamforming on a birds eye view of a city

Resource constrained optimal beam identification in challenging urban environments. Ground truth (red antenna pattern) and predicted optimal beam (yellow line with 95% confidence interval)

 

Energy Savings

We optimize the energy efficiency of 5G networks by scheduling a frequency layer switch-off plan that ensures satisfactory Quality of Service with high confidence.

Energy saving model

Resource constrained optimal beam identification in challenging urban environments.

Radiation Exposure Mitigation

We ensure that 5G users are not exposed to high radiation while minimizing the impact on user experience.

Massive MIMO active antenna

Dynamically adapt the irradiated power (EIRP) over the day to fulfill daily limit while avoiding traffic performance loss.

Project members

Matthew Andrews

Matthew Andrews

Dept Head: Modeling and Optimization Group

Francois Durand

Francois Durand

Research Engineer

Ryo Koblitz

Ryo Koblitz

Senior Research Scientist

APA style publications

  • Lorenzo Maggi, Ryo Koblitz, Qiping Zhu, Matthew Andrews "Tracking the Best Beam for a Mobile User via Bayesian Optimization" IEEE Vehicular Technology Conference (2023) [pdf]

  • Lorenzo Maggi, Claudiu Mihailescu, Qike Cao, Alan Tetich, Saad Khan, Simo Aaltonen, Ryo Koblitz, Maunu Holma, Samuele Macchie, Maria Elena Ruggieri, Igor Korenev, Bjarne Klausen. "Energy Savings under Performance Constraints via Carrier Shutdown with Bayesian Learning" (2023) [pdf]

  • Junghoon Kim, Matthew Andrews, "Learning-Based Adaptive User Selection in Millimeter Wave Hybrid Beamforming Systems", 2023 IEEE International Conference on Communications (ICC) [pdf]

  • Lorenzo Maggi, Alvaro Valcarce, and Jakob Hoydis "Bayesian Optimization for Radio Resource Management: Open Loop Power Control" IEEE Journal on Selected Areas in Communications (2021). doi: 10.1109/JSAC.2021.3078490 [pdf]