Using SmartPhone with Un-scaled Map for Indoor Localization


The pedestrian dead reckoning (PDR) system is a selfcontained technique that has been proposed for indoor localization. It only requires a few sensors to be put on the user and can be used for any building without pre-installed beacon nodes or pre-built RF maps/propagation models based on surveys of the environment. Most PDR systems use inertial sensors (such as accelerometer, gyroscope, and digital compass) to measure step count/length and heading direction. Some of the prior work also combined other sensors, e.g., stereo vision [1], ultrasound [2], RFID [3], and laser range scanner [4], to calibrate their results. However, the PDR system has a wellknown problem called sensor drift that results in an inaccurate estimation of the distance and direction when the sensor drift error is accumulated over a period of time. Some prior work [5, 6] utilized the building map with detailed scales information, as shown in Figure 1(a), to calibrate the sensor drift for indoor localization. However, in practice, such a scaled map is difficult to obtain. In this work, we propose a methodology that uses an un-scaled map, as
shown in Figure 1(b), which is generally easier to find, to calibrate the sensor drift. The challenge in using an un-scald map lies in how to match the user trajectory with the map information.sensor readings from a smart phone and use the sensor information to construct the trajectory of the user. Next, we transform the un-scaled map into a link-node model [7] from
which we can get information like the angles of turning corners using Rotation Matrix. We assume that the initial position of the user is known (e.g. at the entrance of a building). We can
then create a list of all possible routes on the map from the initial position. Finally, a set of filters are employed to find the most ‘similar’ route on the map to the user’s trajectory data. In
addition, when RSSI information is available (e.g. there are existing APs in the building), our system can also utilize such information to further improve the accuracy of our results.



WEN-YUAH SHIH, KUN-CHAN LAN, "Using SmartPhone with Un-scaled Map for Indoor Localization," 9th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON), 2012, 18-21 June 2012.


TITLE = {Using SmartPhone with Un-scaled Map for Indoor Localization},
BOOKTITLE = {IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON)},
MONTH = {June},
YEAR = {2012}


pdf Full Text (PDF Format)