• español
    • English
  • Login
  • español 
    • español
    • English
  • Tipos de Publicaciones
    • bookbook partconference objectdoctoral thesisjournal articlemagazinemaster thesispatenttechnical documentationtechnical report
Ver ítem 
  •   IMDEA Networks Principal
  • Ver ítem
  •   IMDEA Networks Principal
  • Ver ítem
JavaScript is disabled for your browser. Some features of this site may not work without it.

Cleaning Matters! Preprocessing-enhanced Anomaly Detection and Classification in Mobile Networks

Compartir
Ficheros
Paper (1.613Mb)
Identificadores
URI: http://hdl.handle.net/20.500.12761/1593
Metadatos
Mostrar el registro completo del ítem
Autor(es)
Ramirez, Juan Marcos; Díez Muñoz, Fernando; Rojo, Pablo; Mancuso, Vincenzo; Fernández Anta, Antonio
Fecha
2022-06
Resumen
Mobile communications providers often monitor key performance indicators (KPIs) with the goal of identifying anomalous operation scenarios that can affect the quality of Internet-based services. In this regard, anomaly detection and classification in mobile networks has become a challenging task due to the unknown distributions exhibited by the collected data and the lack of interpretability of the embedded (machine learning) models. This paper proposes an unsupervised end-to-end methodology based on both a data cleaning strategy and explainable machine learning models to detect and classify performance anomalies in mobile networks. The proposed approach, dubbed clean and explainable anomaly detection and classification (KLNX), aims at identifying attributes and operation scenarios that could induce anomalous KPI values without resorting to parameter tuning. Unlike previous methodologies, the proposed method includes a data cleaning stage that extracts and removes experiments and attributes considered outliers in order to train the anomaly detection engine with the cleanest possible dataset. Additionally, machine learning models provide interpretable information about features and boundaries describing both the normal network behavior and the anomalous scenarios. To evaluate the performance of the proposed method, a testbed generating synthetic data is developed using a known TCP throughput model. Finally, the methodology is assessed on a real data set captured by operational tests in commercial networks.
Compartir
Ficheros
Paper (1.613Mb)
Identificadores
URI: http://hdl.handle.net/20.500.12761/1593
Metadatos
Mostrar el registro completo del ítem

Listar

Todo IMDEA NetworksPor fecha de publicaciónAutoresTítulosPalabras claveTipos de contenido

Mi cuenta

Acceder

Estadísticas

Ver Estadísticas de uso

Difusión

emailContacto person Directorio wifi Eduroam rss_feed Noticias
Iniciativa IMDEA Sobre IMDEA Networks Organización Memorias anuales Transparencia
Síguenos en:
Comunidad de Madrid

UNIÓN EUROPEA

Fondo Social Europeo

UNIÓN EUROPEA

Fondo Europeo de Desarrollo Regional

UNIÓN EUROPEA

Fondos Estructurales y de Inversión Europeos

© 2021 IMDEA Networks. | Declaración de accesibilidad | Política de Privacidad | Aviso legal | Política de Cookies - Valoramos su privacidad: ¡este sitio no utiliza cookies!