logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-09162025-175614


Tipo di tesi
Tesi di laurea magistrale
Autore
DI BELLA, DANIELE
URN
etd-09162025-175614
Titolo
An eBPF-Extended 5G UPF for Packet Processing and Observability
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
CYBERSECURITY
Relatori
relatore Prof. Lettieri, Giuseppe
relatore Prof. Procissi, Gregorio
relatore Dott.ssa Fais, Alessandra
Parole chiave
  • 5G
  • eBPF
  • network monitoring applications
  • programmable networks
  • software data plane
  • UPF
Data inizio appello
02/10/2025
Consultabilità
Non consultabile
Data di rilascio
02/10/2028
Riassunto
In recent years, many research efforts has focused on extending the capabilities of switches by adding advanced monitoring features. In this thesis, we argue that a better vantage point for monitoring activities is the User Plane Function (UPF) available in the 5G core network architecture. This thesis will explore a new conception of an advanced UPF that exposes a programmable interface for monitoring applications and services. In this way, the UPF will act as a new hook-point for eBPF-based monitoring functionalities. For their part, the applications and services will follow the Stream Processing paradigm, with WindFlow playing a central role, and will range from basic queries, such as flow-meters and heavy-hitter detectors, to more complex traffic probes and in-packet processors.
File