Indoor positioning system

• Grant number: 19/00850-0
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: May 01, 2020
• End date: April 30, 2021
• Field of knowledge: Physical Sciences and Mathematics - Mathematics - Applied Mathematics
• Principal Investigator: Alexandre Picchi Neves
• Grantee: Alexandre Picchi Neves
• Company: Lumminy Consultoria e Serviços em Informática Ltda
• CNAE: Desenvolvimento e licenciamento de programas de computador customizáveis ; Consultoria em tecnologia da informação ; Atividades de monitoramento de sistemas de segurança
• City: Indaiatuba
• Pesquisadores principais: Francisco de Assis Magalhães Gomes Neto
• Associated scholarship(s): 20/05802-1 - Indoor positioning system, BP.TT; 20/06436-9 - Indoor positioning system, BP.TT

Abstract

The GPS (Global Positioning System) is the most used method for geolocation in open areas. However, for indoor environments, such as industries and hospitals, this model is not applicable due to satellite reach limitations and signal losses. In the first stage of this project, some technologies for indoor positioning were analyzed, such as Wi-Fi, RFID (radio frequency identification), Bluetooth, NFC (near field communication), ultrawide band, and infrared and magnetic sensors. Some of these technologies did not show satisfactory accuracy, and were discarded. Others, like active badges, are expensive and hard to use. On the other hand, with its low cost, ease of implementation and good resolution, Wi-Fi technology adapts well to locating moving people indoors. Among the Wi-Fi based location strategies, we have chosen the one based on the received signal strength indicator (RSSI). The RSSI value represents the signal strength transmitted by an access point (AP), that loses force as it propagates, i.e., the closer to the access point, the better the signal strength. However, wireless signal strength is also influenced by factors such as receiver sensitivity, attenuation, path loss and the presence of other signals in the environment. The methodology used consists in capturing the power of the signal from strategically located antennas and use it as a fingerprint for the location of the sensor, according to its particular environmental characteristics (distance to the AP, signal strength, temperature, humidity, noise etc.). The fingerprint method is divided into two phases. In the first one, called the "calibration phase", the information at each point of a given environment is collected, converted into a unique fingerprint (that is generated by a deterministic or probabilistic algorithm), and stored in a database.In the second phase, the location of an employee is estimated according to its position and the previously captured data. The information collected by the sensor is used to create a fingerprint that is compared with the data stored in the database. The actual position of the employee is given by the point for which the fingerprint is closest to the one generated by the sensor. (AU)