Dept. of Communication Engineering, Yadegar -e- Imam Khomeini (RAH) Branch, Islamic Azad University, Tehran, Iran
: Received: 27 July 2016 / Approved: 27 July 2016 / Online: 27 July 2016 (10:38:46 CEST)
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
How to cite:
Karegar, P. Wireless Sensor Network Fingerprinting Localization Using Affinity Propagation Technique. Preprints2016, 2016070084 (doi: 10.20944/preprints201607.0084.v1).
Karegar, P. Wireless Sensor Network Fingerprinting Localization Using Affinity Propagation Technique. Preprints 2016, 2016070084 (doi: 10.20944/preprints201607.0084.v1).
Due to the enlarging advancement of wireless sensor network localization, indoor localization using fingerprint has become more prominent in recent years. It encompasses a database called Receive Strength Signal Indicator (RSSI) vectors, which is a primitive quantity in wireless sensor network fingerprinting localization. The equivalence of some methods is pointed out from the literatures, and some new variants are presented in this study. First affinity propagation is used for clustering data points in the offline phase, next the online phase localization algorithms is exploited. It entails two stages coarse localization and fine localization. In the coarse localization step, both metric of similarity to the receive strength signal vector of exemplars, and resemblance to the weighted average receive strength signal vector of the cluster members are applied. In online phase, both deterministic and probabilistic algorithms are evaluated. Moreover the impact of preference variation within affinity propagation clustering will be investigated in accuracy and real-time ability. Ultimately, two coarse localization methods as Mahalanobis norm method and similarity to exemplar receive strength signal vector are compared based on positioning accuracy and performance. Experimental outcomes prove that our prospective algorithm will promote the accuracy and localization error compared with the method without clustering.
Localization using fingerprint, affinity propagation, coarse localization, fine positioning, Receive Strength Signal