Optimum NN Algorithms Parameters on the UJIIndoorLoc for Wi-Fi Fingerprinting Indoor Positioning Systems

Ebaid, Emad and Navaie, Keivan (2023) Optimum NN Algorithms Parameters on the UJIIndoorLoc for Wi-Fi Fingerprinting Indoor Positioning Systems. In: 2022 32nd International Telecommunication Networks and Applications Conference (ITNAC) :. IEEE, NZL, pp. 280-286. ISBN 9781665471046

[thumbnail of Optimum_NN_Algorithms_Parameters_on_the_UJIIndoorLoc_for_Wi-Fi_Fingerprinting_Indoor_Positioning_Systems_Ebaid & Navaie]
Text (Optimum_NN_Algorithms_Parameters_on_the_UJIIndoorLoc_for_Wi-Fi_Fingerprinting_Indoor_Positioning_Systems_Ebaid & Navaie)
Optimum_NN_Algorithms_Parameters_on_the_UJIIndoorLoc_for_Wi_Fi_Fingerprinting_Indoor_Positioning_Systems_Ebaid_Navaie.pdf - Accepted Version
Available under License Creative Commons Attribution-NonCommercial.

Download (490kB)

Abstract

Wi-Fi fingerprinting techniques are commonly used in Indoor Positioning Systems (IPS) as Wi-Fi signal is available in most indoor settings. In such systems, the position is estimated based on a matching algorithm between the enquiry points and the recorded fingerprint data. In this paper, our objective is to investigate and provide quantitative insight into the performance of various Nearest Neighbour (NN) algorithms. The NN algorithms such as KNN are also often employed in IPS. We extensively study the performance of several NN algorithms on a publicly available dataset, UJIIndoorLoc. Furthermore, we propose an improved version of the Weighted KNN algorithm. The proposed model outperforms the existing works on the UJIIndoorLoc dataset and achieves better results for the success rate and the mean positioning error.

Item Type:
Contribution in Book/Report/Proceedings
ID Code:
194482
Deposited By:
Deposited On:
26 May 2023 13:40
Refereed?:
Yes
Published?:
Published
Last Modified:
29 Jun 2024 01:00