RSSI/IMU sensor fusion-based localization using unscented Kalman filter

The most crucial problem in navigation system is localization. The global positioning system (GPS) has long been used in mobile unit localization. However, GPS is incapable in some situation such as indoor environment. The received signal strength indicator (RSSI) from wireless communication is a promising alternative method to derive the location of a mobile unit. To improve the precision and robustness in the GPS-based localization, an inertial measurement unit (IMU) is normally used. In this paper, we study the possibility to use RSSI from wireless local area network (WLAN) and an IMU to derive the location of the mobile unit. We apply unscented Kalman filter (UKF) as a fusion engine for those two information. The experiment is conducted by using mobile unit equipped with low-cost IMU and a wireless communication module together with multiple access points to evaluate the performance of our algorithm, and the result is promising.