1. CONICET
  2. Producciónes CyT
Producción CyT
Advanced Constellation Emulation and Synthetic Datasets Generation for Non-Terrestrial Networks

Artículo

Autoría
FRAIRE, JUAN ANDRES
Fecha
2024
Editorial y Lugar de Edición
Institute of Electrical and Electronics Engineers Inc.
Revista
2024 IEEE International Mediterranean Conference on Communications and Networking, MeditCom 2024 (pp. 37-43) Institute of Electrical and Electronics Engineers Inc.
Resumen Información suministrada por el agente en SIGEVA
Mega satellite constellations, now realized entities,encompass thousands of nodes. However, efficient orchestrationof multi-hop paths and distributed processing tasks in Non-Terrestrial Networks (NTN) remains a considerable challenge.The integration of NTN systems into 5G cellular networks neces-sitates innovative adaptations of Software-Defined Networking(SDN) and Multi-access Edge Computing (MEC) to suit thedynamic environments of NTN. In this context, we presentMeteorNet, a state-of-the-art ... Mega satellite constellations, now realized entities,encompass thousands of nodes. However, efficient orchestrationof multi-hop paths and distributed processing tasks in Non-Terrestrial Networks (NTN) remains a considerable challenge.The integration of NTN systems into 5G cellular networks neces-sitates innovative adaptations of Software-Defined Networking(SDN) and Multi-access Edge Computing (MEC) to suit thedynamic environments of NTN. In this context, we presentMeteorNet, a state-of-the-art emulation tool conceived for satelliteconstellations. MeteorNet accurately replicates the behavior ofNTNs by implementing space orbits, Earth rotation calculations,and Linux network interfaces across diverse network layers.Coupled with a continuous measurement system founded onsFlow, MeteorNet compiles critical switch variables in a cen-tralized database, thus providing a distinctive methodology forcreating realistic synthetic datasets. The pertinence of syntheticdatasets is paramount in NTN, given the scarcity of operativesystems and the inaccessibility of accurate data from the fewexisting systems due to proprietary constraints. These datasetsare instrumental for formulating and training intelligent controlalgorithms and Machine Learning (ML) models for SDN andMEC advancements in NTN. To illustrate the efficacy of thisapproach, we explore a realistic networking case study with aring topology, demonstrating how data models describe intricaterouting and edge computing protocols for NTN.
Ver más Ver menos
Palabras Clave
Space Cloud