Comparative Analysis of Partition-based and Spectral Clustering for Satellite Geometry Optimization in Combined GPS and NavIC ConstellationsComparative Analysis of Partition-based and Spectral Clustering for Satellite Geometry Optimization in Combined GPS and NavIC Constellations
Disaster Advances · 31.07.2026 · 19 (9) · 71
DOI: 10.25303/199da071081 ↗Kurz gefasst
Fachartikel in der Zeitschrift Disaster Advances, online erschienen am 31.07.2026. Verfasst von Modhini Buri und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Satellite geometry significantly influences the positioning performance of Global Navigation Satellite Systems (GNSS). An uneven spatial distribution of visible satellites results in increased Dilution of Precision (DOP), which directly degrades positioning performance. This study presents a comparative analysis of partition-based and graph-based (spectral) clustering techniques for improving satellite geometry in an integrated GPS-NavIC multi-constellation system. Partitioning methods such as K-means and kmedoids are compared with spectral clustering approaches using different similarity measures including Hamming, Euclidean and Chebyshev distances. The objective is to select an optimal subset of visible satellites that minimizes geometric dilution of precision (GDOP), positional dilution of precision (PDOP), horizontal dilution of precision (HDOP) and vertical dilution of precision (VDOP). Satellite visibility information is extracted from RINEX observation data acquired at the GITAM University IGS receiver, Visakhapatnam, India. Experimental results show that spectral clustering with Hamming distance, combined with medoid-based satellite selection, yields the lowest GDOP values and enhances position accuracy relative to conventional partition-based clustering methods. The findings demonstrate the effectiveness of spectral clustering techniques for satellite selection and geometry optimization in multi-constellation GNSS environments.
Zitieren
APA
Buri, M., & V.B.S., S. I. D. (2026). Comparative Analysis of Partition-based and Spectral Clustering for Satellite Geometry Optimization in Combined GPS and NavIC ConstellationsComparative Analysis of Partition-based and Spectral Clustering for Satellite Geometry Optimization in Combined GPS and NavIC Constellations. Disaster Advances, 19(9), 71. https://doi.org/10.25303/199da071081
Vancouver
Buri M, V.B.S. SID. Comparative Analysis of Partition-based and Spectral Clustering for Satellite Geometry Optimization in Combined GPS and NavIC ConstellationsComparative Analysis of Partition-based and Spectral Clustering for Satellite Geometry Optimization in Combined GPS and NavIC Constellations. Disaster Advances. 2026;19(9):71. doi:10.25303/199da071081
BibTeX
@article{Buri2026,
author = {Buri, Modhini and V.B.S., Srilatha Indira Dutt},
title = {Comparative Analysis of Partition-based and Spectral Clustering for Satellite Geometry Optimization in Combined GPS and NavIC ConstellationsComparative Analysis of Partition-based and Spectral Clustering for Satellite Geometry Optimization in Combined GPS and NavIC Constellations},
journal = {Disaster Advances},
year = {2026},
volume = {19},
number = {9},
pages = {71},
doi = {10.25303/199da071081}
}
EUVIMED