An Empirical Study on the Occupancy Detection Techniques Based on Context-Aware IoT System

Abstract

Occupancy detection and behavior in buildings has a huge impact on cooling, heating, ventilation demand, building controls, and energy consumption in lighting appliances. The human factor is an important factor in real-time occupancy information and building energy management systems that offer great potential for maximizing energy efficiency and assessing energy flexibility. The occupancy predictive strategy provided a better quality of service and energy savings performance than reactive strategies. In this research paper, 20 papers based on context-aware IoT systems for occupancy detection are reviewed. The research works are categorized into the sensor, sensor fusion, Wi-Fi, LAN, radio frequency (RF) signals, machine learning, and so on. The research gaps and the challenges faced during the occupancy detection are listed for further enhancement in the occupancy detection methods. The research work is analyzed based on the performance metrics, classification methods, and the publication year. The analysis shows that the most frequently used performance metrics is accuracy, the most commonly used classification technique is the sensor, whereas most of the research papers are published in the year 2018.