Database Optimization of Fingerprint-Based Indoor Positioning System Using Genetic Algorithm

Abstract

Indoor positioning systems are becoming more and more popular nowadays. There are many challenges in designing such systems. An effective method of designing these systems is to employ Wi-Fi technology along with the fingerprinting algorithm. This algorithm consists of an offline or setup phase and an online or exploitation phase. A challenge that these systems face is the offline phase, in which a database is collected from the signal intensities of modems existing at different points in an environment. In this case, the large volume of the database demands high rates of temporal costs and human labor, which increases the setup costs of these systems. At the same time, decreasing the number of sampling points will reduce the positioning accuracy. The genetic algorithm was used in this paper to select and arrange the number of database points in order to decrease the database volume and keep accuracy within an acceptable range.