IEEE_ComLetters_MRA_2023 preprint.pdf (433.53 kB)

Transform-based Multiresolution Decomposition for Unsupervised Learning and Data Clustering of Cellular Network Behaviour.

Download (433.53 kB)
posted on 2023-07-17, 03:04 authored by Juan Cantizani-EstepaJuan Cantizani-Estepa, Sergio FortesSergio Fortes, Javier Villegas, Javier Rasines Suarez, Raúl Martín Cuerdo, Eduardo Baena, Raquel Barco

The growing complexity of cellular networks makes it harder for network operators to monitor and manage the system. To ease the management and automatically detect network problems, unsupervised techniques have been put to use. This work proposes a novel method that combines Multi-Resolution Analysis (MRA) by wavelet transforms and unsupervised clustering for the totally unsupervised grouping of cellular network behaviours through different Key-Performance Indicator (KPI)s. The application of multi-resolution decomposition, allows the much simpler clustering technique to take into account temporal information that would require of a much complex method otherwise. The proposed approach has been tested with real network data successfully separating different behaviours.


Ref.- 8.06/5.59.5705 -3 IDEA, “Desarrollo de casos de uso para el diseño, optimización y dimensionado de redes móviles - Líneas B1 y D1”, as part of the Agency IDEA incentives.

Grant FPU21/04472 from the “Ministerio de Ciencia e Innovación”

Project MAORI from the Spanish government “Ministerio de Asuntos Económicos y Transformación Digital y la Unión Europea - NextGenerationEU”, inside the plan entitled “Plan de Recuperación, Transformación y Resiliencia y el Mecanismo de Recuperación y Resiliencia”.

University of Málaga by the “II Plan Propio de Investigación y Transferencia de la Universidad de Málaga”.


Email Address of Submitting Author

ORCID of Submitting Author


Submitting Author's Institution

Universidad de Málaga

Submitting Author's Country

  • Spain