A Low-Cost Indoor Localization System for Energy Sustainability in Smart Buildings

This paper proposes a low-cost indoor localization system in the context of smart buildings in which customized services are provided to occupants, considering energy consumption issues. To offer indoor services, it is necessary to consider information concerning to the identity and location of the occupants, as well as the accuracy required in the location data to provide these services in a comfortable and energy efficient way. We propose a mechanism for solving these localization requirements, using radio frequency identification and infrared data. The computational techniques implemented in this paper in order to solve the indoor localization problem are: 1) a radial basis function (RBF) network, which is in charge of estimating the user position and 2) a particle filter, which is in charge of estimating the next position of users based on the previous positions estimated by the RBF. This localization mechanism is tested in a reference smart building that houses an automation system for collecting data and controlling the devices and appliances of the building. The results obtained from the assessments performed are very satisfactory in terms of accuracy and precision of the user location data, providing a low-cost solution for ambient adaptation based on human presence.