Applying A New Localization Approach Based On Multidimensional Wifi Fingerprinting
1382 Words6 Pages
￼A fine-grained Indoor Localization using Multidimensional WiFi Fingerprinting
Abstract—Although fingerprint based localization is promising for indoor applications, its accuracy remains a huge challenge. Most existing approaches depends on the Radio Signal Strength (RSS) to generate fingerprints. Merely using RSS is unable to accurately localize since such an one-dimensional fingerprint is serious influenced by the interference and multipath effect in the indoor environment. In this paper, we propose a new localization approach based on multidimensional WiFi fingerprint. Instead of using RSS only, we employ RSS, transmitted power, and channel information to construct fingerprints. The extension of fingerprints enables fine-grained localization and tracking services. We also deign a cosine similarity based matching algorithm and enhanced particle filter mechanism to realize the localization and tracking. Extensive experiment and implementation results show that using the new fingerprint can achieve an accuracy within 2m in 90% testing points, with good adaptability to complex indoor environments.
Keywords—Indoor positioning; Multidimensional fingerprint; WIFI. Fingerprint matching
Accurate indoor localization is crucial for many applications, such as the navigation, tour guiding, escaping, etc. Different from outdoor localization, which mainly depends on the global positioning system (GPS), providing accurate positioning service is difficult in the indoor