Indoor Positioning and Fingerprint Updating Based on Affinity Propagation Clustering

In recent years, positioning in the indoor WiFi scene, the location of the fingerprint method is used in wireless indoor positioning, there are already through the smart phone automatically capture RSS fingerprints to build fingerprint library, reducing the workload of offline stage, but the automatic collection of RSS has a certain degree of unreliability and the updating of the fingerprint database is not timely and so on. In this paper, the automatic collection of fingerprints based on affinity propagation clustering, remove singular values, detect AP update method to ensure the reliability of fingerprint database. In the online phase WKNN and APC algorithms are combined. Experiments show that the improved fingerprint library can improve the positioning effect effectively by 20%, and make the automatic collection data based on the general package have good robustness