Dr. Yaya Etiabi

Dr. Yaya Etiabi

He/Him/His

Ph.D., Eng. | Computer & Communication Sciences and Engineering | Researcher: ISCC & Edge AI, IoT| AI Literacy Advocate

Talks and presentations

Deep Reinforcement Learning for LoRa Localization

June 23, 2023

Talk, IEEE 97th Vehicular Technology Conference (VTC2023-Spring), Piazza Adua 1, Florence, Italy

This talk presents a deep reinforcement learning (DRL) solution for LoRa localization. It addresses the challenges of accurate localization with LoRa technology. Specifically, I presented a novel approach that uses DRL to optimize the assignment of spreading factors (SFs) in LoRa networks to improve localization accuracy. The results of the research demonstrate that the DRL-based approach achieves superior localization accuracy compared to benchmark methods.

Federated Learning for Hierarchical 3D Indoor Localization

March 29, 2023

Talk, IEEE Wireless Communications and Networking Conference (WCNC), Glasgow, Scotland, UK

In this presentation, I introduced a federated learning (FL) framework for hierarchical 3D indoor localization using a deep neural network. This work tackles the challenges of privacy issues and resource constraints in indoor positioning. I presented an approach that exploits the hierarchy between floors and buildings to train an accurate localization model while preserving user privacy.

Huber Estimator and Statistical Bootstrap for IoT Localization

May 22, 2021

Talk, UNet 2021 - International Conference on Ubiquitous Networking, Marrakesh, Morocco [Virtual]

This presentation details a novel approach to improve the accuracy of localization in IoT systems. I presented a light-weight localization method using the Huber estimator and statistical bootstrapping. The method refines position estimates in a two-step process to enhance localization accuracy in IoT environments.

Distributed and Collaborative Localization for IoT

November 30, 2020

Talk, 18th International Conference on Advances in Mobile Computing & Multimedia, Chiang Mai, Tailand [Virtual]

In this talk, I presented a distributed localization algorithm designed for IoT environments. The algorithm is based on a convex relaxation of the Huber loss function. I discussed how this approach addresses the challenges of achieving accurate localization in IoT networks, particularly in the presence of Non-Line-of-Sight links.