Shagufta Henna - Research

Projects

Graph Neural Networks for Virtualized Networks for Resource Management , TIDA, Science Foundation of Ireland

The main objectives of the project are to develop a commercial prototype system based on neural networks integrated into the standardized Openstack/ONAP and validate the machine learning-based system in an industry relevant environment in collaboration with an Industry partner (TRL 5). Finally, in this project we aim to determine a viable route to market along with a credible business case to commercialize.

Affinity-based Neural Network Approach with Attention for High Density Wireless Local Area Networks

In this proposal, in the first instance, we plan to investigate the most critical affinity rules from an operator ‘s perspective with a focus on Cisco's equipment. Based on these affinity rules, we aim to develop a novel deep neural network, which can predict the most suitable configuration of existing APs based on the affinity/anti-affinity rules from an existing indoor WLAN deployment.

Investigation on hybrid RF/FSO links for dust storm conditions

In this project, a cognitive radio mechanism is investigated to employ the intelligent switch using Fuzzy Logic and Deep Reinforcement learning to ensure switching between two signals efficiently according to the weather and atmospheric conditions. The system is implemented based on the environment of the optical communication and RF communication. (Co. Investigator)

ITU Recommendation Focus Group on Machine Learning for 5G and Future Generation Networks

"Requirements,architecture and design for machine learning function orchestrator", International Telecommunication Union (ITU) , Focus Group on Machine Learning for Future Networks including 5G specification.

RECENT TALKS

7^th March 2019. Distributed Deep Learning for the Next Generation High Density Wireless Local Area Networks (WLANs) Deployments, organized by the China Academy of Information and Communications Technology (CAICT) in collaboration with International Telecommunication Union (ITU) in Shenzhen.
7^th March 2019. Automated Machine Learning for Future Networks including 5G (Auto5GML), organized by the China Academy of Information and Communications Technology (CAICT) in collaboration with ITU in Shenzhen.
2^nd June 2020. Requirements and Architecture for the Machine Learning Framework Orchestrator, organized by the ITU, Geneva, Switzerland.
31^st July 2020. ITU-T Global Challenge on the Reference Implementation of Machine Learning Function Orchestrator, Organized by ITU, Geneva, Switzerland.