Authors: Rafael A. Aguiar, Nuno Paulino, Luís M. Pessoa

Published on: May 03, 2024

Impact Score: 7.2

Arxiv code: Arxiv:2405.01928

Summary

• What is new: Introduction of two machine learning optimization algorithms that significantly enhance position estimation in RIS-aided localization for Non-Line-of-Sight conditions.
• Why this is important: Improving accuracy in position estimation for mobile user equipment in Non-Line-of-Sight conditions.
• What the research proposes: Two methods leveraging machine learning optimization algorithms capable of achieving sub-centimeter or sub-millimeter level accuracy.
• Results: Significant improvements in indoor mobile localization accuracy, with errors under 30 cm in 90% of cases, and under 5 mm in close to 85% of cases in a simulated room environment.

Technical Details

Technological frameworks used: nan

Models used: Genetic Algorithm (GA), Particle Swarm Optimization (PSO)

Data used: Simulation data for indoor localization in a 10m x 10m room with RIS tiles on two walls.

Potential Impact

Companies in the indoor navigation, smart buildings, and IoT sectors could benefit or be disrupted by these findings.

Want to implement this idea in a business?

We have generated a startup concept here: LocPrecision.