Smartphone Based Indoor Position Estimation

Positioning or location estimation systems are adopted widely in modern world, to locate objects or users. Current broadly adopted positioning sys- tems such as global positioning system (GPS) or global navigation satellite system (GNSS) are limited to outdoor applications and provide poor locating accuracy for indoor positing with short movement span. Through this study we aim to evaluate the possibility of utilizing received signal strength indicator (RSSI) value of a smartphone antenna with the com- bination of trilateration theory and feed-forward back-propagation neural network (FFBPNN) to identify the correct indoor location of a person in two-dimensional (2D) space. This research area has recently attracted a great deal of attention especially in assisted living applications. For the evaluation of final outcome, we have selected one of the most powerful statistical accuracy measure; mean absolute percentage error (MAPE). The results show that the proposed indoor positioning system has 73.4% of accuracy which can provide acceptable localization accuracy as same as the traditional approaches.