Abstract
Recently location based services (LBS) have become increasingly popular in indoor environments. Among these indoor positioning techniques providing LBS, a fusion approach combining WiFi-based and pedestrian dead reckoning (PDR) techniques is drawing more and more attention of researchers. Although this fusion method performs well in some cases, it still has some limitations, such as heavy computation and inconvenience for real-time use. In this work, we study map information of a given indoor environment, analyze variations of WiFi received signal strength (RSS), define several kinds of indoor landmarks, and then utilize these landmarks to correct accumulated errors derived from PDR. This fusion scheme, called Landmark-aided PDR (LaP), is proved to be light-weight and suitable for real-time implementation by running an Android application designed for the experiment. We compared LaP with other PDR-based fusion approaches. Experimental results show that the proposed scheme can achieve a significant improvement with an average accuracy of 2.17 m.
Highlights
IntroductionIndoor mobile positioning techniques are the backbone of location based services (LBS)
Location based services (LBS) are becoming increasingly popular in indoor environments because massive wireless networks are built according to the IEEE 802.11 wireless Ethernet standard.Indoor mobile positioning techniques are the backbone of location based services (LBS)
We focus on the positioning accuracy among Landmark-aided PDR (LaP) and other multi-fusion approaches
Summary
Indoor mobile positioning techniques are the backbone of LBS. These techniques can be generally divided into two categories according to different measurements adopted: pedestrian dead reckoning (PDR) based on inertial sensors such as accelerometers, gyroscopes, etc. [1]; and location determination employing received signal strength (RSS) of WiFi as a metric [2]. PDR is a self-contained approach but will produce a growing drift as walking distance increases [3]. It relies on readings of inertial sensors embedded in smartphones to detect steps, calculate step length and determine walking direction. RSS-based positioning mainly includes the model-based approach and fingerprinting method. The fingerprinting method has a higher accuracy, but requires tedious
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
