6533b826fe1ef96bd128465d

RESEARCH PRODUCT

Positioning Error Prediction and Training Data Evaluation in RF Fingerprinting Method

Tapani RistaniemiSyed KhandkerRiaz Mondal

subject

SIMPLE (military communications protocol)Computer sciencecomputer.internet_protocolmedia_common.quotation_subjectQuality of service020206 networking & telecommunications02 engineering and technologyFingerprint recognitioncomputer.software_genrePosition (vector)Location-based service0202 electrical engineering electronic engineering information engineeringRADIUS020201 artificial intelligence & image processingQuality (business)Radio frequencyData miningcomputermedia_common

description

Radio Frequency (RF) fingerprinting-based localization has become a research interest due to its minimum hardware requirement and satisfiable positioning accuracy. However, despite the significant attention this topic has gained, most of the research focused on the calculation of position estimates. In this paper, we propose a simple and novel method that can be used as an indicator of fingerprinting positioning error. The method is based on cluster radius evaluation of multiple fingerprinting data during the test phase, which can be used by a Location Based Service (LBS) provider to predict the user position estimation accuracy. This method can be used effectively in real-time to predict the estimation error and thereby assists the LBS to offer a better quality of service. The cluster radius also reveals the quality of the recorded radio map.

yearjournalcountryeditionlanguage
2019-09-012019 International Conference on Indoor Positioning and Indoor Navigation (IPIN)
https://doi.org/10.1109/ipin.2019.8911821