Machine Learning Approaches for Indoor Positioning: A Case Study on KNN vs WKNN-inv and WKNN-cos
Gharyan University Journal of Engineering Science · 15.03.2026 · 2 (1) · 1-10
DOI: 10.65568/gujes.2026.020101 ↗Kurz gefasst
Fachartikel in der Zeitschrift Gharyan University Journal of Engineering Science, online erschienen am 15.03.2026. Verfasst von Mohammed Aboughlia und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref).
Abstract (Originalsprache)
Indoor localization is of paramount importance in intelligent environments, particularly in scenarios where Global Positioning System (GPS) signals are hindered by physical barriers within indoor settings. This study investigates machine learning methodologies designed to enhance the precision of indoor positioning systems. In particular, we conduct a comparative analysis of three algorithms: the conventional KNearest Neighbors (KNN), a Weighted KNN utilizing inverse distance metrics (WKNN-inv), and an additional variant that employs cosine similarity measures (WKNN-cos). Each algorithm was systematically implemented and assessed concerning its positioning accuracy. The results from simulations indicate that WKNN-inv yields an accuracy enhancement of approximately 11.2% relative to KNN, whereas WKNN-cos provides a 13.0% improvement.
Zitieren
APA
Aboughlia, M., & Elalem, M. (2026). Machine Learning Approaches for Indoor Positioning: A Case Study on KNN vs WKNN-inv and WKNN-cos. Gharyan University Journal of Engineering Science, 2(1), 1-10. https://doi.org/10.65568/gujes.2026.020101
Vancouver
Aboughlia M, Elalem M. Machine Learning Approaches for Indoor Positioning: A Case Study on KNN vs WKNN-inv and WKNN-cos. Gharyan University Journal of Engineering Science. 2026;2(1):1-10. doi:10.65568/gujes.2026.020101
BibTeX
@article{Aboughlia2026,
author = {Aboughlia, Mohammed and Elalem, Mohamed},
title = {Machine Learning Approaches for Indoor Positioning: A Case Study on KNN vs WKNN-inv and WKNN-cos},
journal = {Gharyan University Journal of Engineering Science},
year = {2026},
volume = {2},
number = {1},
pages = {1--10},
doi = {10.65568/gujes.2026.020101}
}
EUVIMED