WiFi-assisted multi-floor indoor localization with inertial sensors

With the rapid development of Location Based Service (LBS) applications and the continual improvement of intelligent terminal technology, study on smartphone-based indoor positioning approach has turned to be the hot topic quickly in recent years. In this paper, a WiFi-assisted Pedestrian Dead Reckoning (PDR) algorithm based on activity recognition is proposed on the smartphone platform to realize the pedestrian's trajectory tracking and positioning in the multi-floor indoor environment. Our algorithm combines the Inertial Measurement Unit (IMU) data with the Received Signal Strength Indication (RSSI) data from the smartphone to distinguish pedestrian's activities. Then, based on the recognition result, we implement adjustment and optimization on the traditional PDR algorithm. The test results show that our algorithm performs with better accuracy and reliability compared with the traditional algorithm on realistic scenario.