Probability Distribution-Aided Indoor Positioning Algorithm Based on Affinity Propagation Clustering
With the rapid development of indoor positioning technology, the fingerprint-based Wireless Local Area Network (WLAN) positioning becomes a new and widely recognized research concern. This paper proposes a probability distribution-aided indoor positioning algorithm based on the affinity propagation clustering. Different from the conventional fingerprint-based positioning algorithms, our algorithm first uses the affinity propagation clustering to minimize the searching space of reference points (RPs). Then, the probability distribution-aided positioning algorithm is utilized to estimate the target’s accurate position. Furthermore, because the affinity propagation clustering can effectively reduce the computation cost for the RP searching which is involved in the probability distribution-aided positioning algorithm, the positioning efficiency of our proposed algorithm can be effectively guaranteed. Experimental results demonstrate that our proposed affinity propagation clustering will significantly improve the performance of the probability distribution-aided positioning algorithm in both the positioning accuracy and real-time ability.
KeywordsWLAN indoor positioning Fingerprinting Affinity propagation clustering RSS Probability distribution
- 3.Khodayari S, Maleki M, Hamedi E (2010) A RSS-based fingerprinting method for positioning based on historical data. Proc IEEE SPECTS 306–310Google Scholar
- 5.Jain VK, Tapaswi S, Shukla A (2012) Distributed location estimation system using WLAN received signal strength fingerprints. Proc IEEE WCNC 3102–3106Google Scholar
- 7.Wang Y, Li WY, Sun Y (2011) Wireless sensor network cluster locations: a probabilistic inference approach. Proc IEEE ICAL 76–80Google Scholar