Capacity Control in Indoor Spaces Using Machine Learning Techniques Together with BLE Technology

Journal of Sensor and Actuator Networks

M. Encarnación Beato Gutiérrez¹  Roberto Berjón Gallinas¹  Montserrat Mateos Sánchez¹  Ana M. Fermoso García¹ 

[1] Faculty of Computer Science, Universidad Pontificia de Salamanca, 37002 Salamanca, Spain;

关键词: Bluetooth Low Energy (BLE);    machine learning;    COVID capacity control;    level occupancy;    occupancy detection;    indoor detection;   

DOI  :  10.3390/jsan10020035

来源: DOAJ

【 摘 要 】

At present, capacity control in indoor spaces is critical in the current situation in which we are living in, due to the pandemic. In this work, we propose a new solution using machine learning techniques with BLE technology. This study presents a real experiment in a university environment and we study three different prediction models using machine learning techniques—specifically, logistic regression, decision trees and artificial neural networks. As a conclusion, the study shows that machine learning techniques, in particular decision trees, together with BLE technology, provide a solution to the problem. The contribution of this research work shows that the prediction model obtained is capable of detecting when the COVID capacity of an enclosed space is exceeded. In addition, it ensures that no false negatives are produced, i.e., all the people inside the laboratory will be correctly counted.

【 授权许可】

Unknown