A NOVEL FRAMEWORK FOR SMART BUILDING APPLICATIONS IN THE IDENTIFICATION OF VACANT SPACE

Abstract

The intelligent use of building spaces is one of the main demands of the intelligent infrastructure of the present time. The quick advancements in the sphere of smart building technologies and Internet of Things (IoT) systems have allowed constant monitoring of indoor areas, but the correct identification of empty and occupied areas remains a rather problematic issue because of dynamic human behavior and uncertainties in sensors and uneven information sources. The present paper suggests a new paradigm of the smart building application that would help to correctly identify the vacant spaces based on smart data-driven methodology. The framework unites the information gathered through the distributed sensors including motion, occupancy, environmental, and energy usage sensors, with the state-of-the-art machine learning algorithms to examine the real-time and historical trends in space use. Powerful preprocessing and feature extraction algorithms are used to deal with noise, missing, and time variations. The main goal of the suggested framework is to facilitate the quality and reliable detection of the vacant space besides being scalable and energy efficient. Experimental assessments indicate that the framework is efficient in estimating vacant spaces to a high level of accuracy, allowing to optimize the allocation of resources, energy conservation, and comfort of occupants. The suggested solution offers an effective and sensible solution to smart buildings management and sustainable development of infrastructure.