I am a 5G System's Designer at Ericsson Canada where I lead research, early phase standard compliant system designs, algorithm development, and patents for 5G networks. My main research expertise and interests are in developing autonomous and predictive network functions with a strong emphasis on practical solutions. My work draws primarily on methods from stochastic optimization, Bayesian estimation, reinforcement learning, time series forecasting, and deep learning.

I completed my Ph.D. in Electrical and Computer Engineering at Queen's University in 2014, and have worked at Bell Labs and Cisco before joining Ericsson. In my Ph.D. Thesis I proposed a framework for predictive radio resource allocation that forecasts demand and plans content delivery over a time horizon. 

I highly value applied research and university-industry collaborations, and currently lead multiple academic partnerships on intelligence and analytics in future networks  (more details here). 

Research Overview

I am interested in the following research questions:

  1. How can future networks be designed to dynamically scale to the fluctuating data requirements of users/things in the 5G era? In particular I am interested in how anticipatory network functions across various layers of the end-to-end communication infrastructure can be designed.

  2. How can future networks learn to autonomously adapt and optimize their underlying network functions using network and environmental data? While networks do have a plethora of data and feedback mechanisms that can facilitate learning, applying machine intelligence is challenging due to inaccuracies of the measured data, algorithm stability, and real-time implementation constraints. As such, I am interested in identifying what techniques from reinforcement learning, control theory, and robust optimization can be used to develop practical solutions for autonomous networking.

  3. How can networks predict demand and both internal and external factors that affect performance accurately? How can the recent advances in generative artificial intelligence and hybrid deep learning architectures be used to develop individual flow and aggregate spatio-temporal traffic models for 5G networks? And finally, how can these models drive the aforementioned anticipatory and autonomous network functions.

Academic Research Partnerships

  • Machine Intelligence and Big Data Analytics for 5G  Networks and Beyond (NSERC Strategic-Queen's University): Topics of research include network traffic prediction, hybrid deep learning architectures for spatio-temporal mobility modeling, anticipatory edge computing and NFV, predictive mobility-aware scheduling.

  • Autonomous Multi-RAT Radio Resource Management for 5G (OCE ENCQOR): The goal of this project is to investigate how unsupervised machine learning and deep reinforcement learning techniques can be used in dual connectivity management of 5G networks.

  • Sensor-less Sensing of 5G Application Traffic (OCE ENCQOR)The goal of this project is to develop artificial intelligence techniques that can model, classify and predict real-time traffic of 5G applications like virtual reality, autonomous vehicle communication, and massive MTC.


  • We are hiring a Machine Learning Intern in our team at Ericsson to work on novel algorithms for 5G network and radio resource management. Please let me know if you are interested/know a good candidate!

  • The TRL Research Lab at Queen's University is hiring a Postdoctoral Fellow to work on Machine Learning in Future Networks as part of the NSERC Strategic collaboration with Ericsson. If interested please feel free to contact me.

  • Dr. Ramy Atawia was awarded the Queen’s Governor General’s Academic Gold Medal for his Thesis. I had the pleasure of mentoring Dr. Atawia who introduced uncertainty modeling and robustness in predictive video delivery - a great accomplishment Ramy!

© 2018 Hatem Abou-zeid