Study on improvement of fingerprint matching algorithm in wireless LAN based indoor positioning system

By comparing and analyzing several common indoor positioning system solutions, we found that indoor positioning System based on wireless LAN is superior, in which fingerprint matching algorithm was widely used and had a large room to be improved for more precise positioning performance. This paper presents an improved fingerprint matching algorithm, which reduces interference by Gaussian filter, calculates distance more accurately by weighted centroid algorithm. Furthermore, the algorithm improves precision by local prediction model revised by entire prediction model to increase the accuracy of positioning. The proposed algorithm can be widely used in different environment and has strong robustness. Experiments performed on real-world place shows that the result of our system is enhanced more than 26%, as compare to the traditional fingerprint-based WLAN positioning system.