O-RAN is radically shifting how cellular networks are designed, deployed and optimized through network programmability, disaggregation, and virtualization. Specifically, RAN Intelligent Controllers (RICs) can orchestrate and optimize the Radio Access Network (RAN) operations, allowing fine-grained control over the network. RICs provide new approaches and solutions for classical use cases such as on-demand traffic steering, anomaly detection, and Quality of Service (QoS) management, with an optimization that can target single User Equipments (UEs), slices, cells, or entire base stations. Such control can leverage data-driven approaches, which rely on the O-RAN open interfaces to combine large-scale collection of RAN Key Performance Measurements (KPMs) and state-of-the-art Machine Learning (ML) routines executed in the RICs. While this comes with the potential to enable intelligent, programmable RANs, there are still significant challenges to be faced, primarily related to data collection at scale, development and testing of custom control logic for the RICs, and availability of Open RAN simulation and experimental tools for the research and development communities. To address this, we introduce ns-O-RAN, a software integration between a real-world near-real-time RIC and an ns-3 simulated RAN which provides a platform for researchers and telco operators to build, test and integrate xApps. ns-O-RAN extends a popular Open RAN experimental framework (OpenRAN Gym) with simulation capabilities that enable the generation of realistic datasets without the need for experimental infrastructure. We implement it as a new open-source ns-3 module that uses the E2 interface to connect different simulated 5G base stations with the RIC, enabling the exchange of E2 messages and RAN KPMs to be consumed by standard xApps. Furthermore, we test ns-O-RAN with the O-RAN Software Community (OSC) and OpenRAN Gym RICs, simplifying the onboarding from a test environment to production with real telecom hardware controlled without major reconfigurations required. ns-O-RAN is open source and publicly available, together with quick-start tutorials and documentation.

ns-O-RAN. Simulating O-RAN 5G systems in ns-3 / Lacava, Andrea; Bordin, Matteo; Polese, Michele; Sivaraj, Rajarajan; Zugno, Tommaso; Cuomo, Francesca; Melodia, Tommaso. - (2023), pp. 35-44. (Intervento presentato al convegno 15th Workshop on Network Simulator, WNS3 2023 tenutosi a Arlington; USA) [10.1145/3592149.3592161].

ns-O-RAN. Simulating O-RAN 5G systems in ns-3

Andrea Lacava
Primo
;
Francesca Cuomo
Penultimo
;
2023

Abstract

O-RAN is radically shifting how cellular networks are designed, deployed and optimized through network programmability, disaggregation, and virtualization. Specifically, RAN Intelligent Controllers (RICs) can orchestrate and optimize the Radio Access Network (RAN) operations, allowing fine-grained control over the network. RICs provide new approaches and solutions for classical use cases such as on-demand traffic steering, anomaly detection, and Quality of Service (QoS) management, with an optimization that can target single User Equipments (UEs), slices, cells, or entire base stations. Such control can leverage data-driven approaches, which rely on the O-RAN open interfaces to combine large-scale collection of RAN Key Performance Measurements (KPMs) and state-of-the-art Machine Learning (ML) routines executed in the RICs. While this comes with the potential to enable intelligent, programmable RANs, there are still significant challenges to be faced, primarily related to data collection at scale, development and testing of custom control logic for the RICs, and availability of Open RAN simulation and experimental tools for the research and development communities. To address this, we introduce ns-O-RAN, a software integration between a real-world near-real-time RIC and an ns-3 simulated RAN which provides a platform for researchers and telco operators to build, test and integrate xApps. ns-O-RAN extends a popular Open RAN experimental framework (OpenRAN Gym) with simulation capabilities that enable the generation of realistic datasets without the need for experimental infrastructure. We implement it as a new open-source ns-3 module that uses the E2 interface to connect different simulated 5G base stations with the RIC, enabling the exchange of E2 messages and RAN KPMs to be consumed by standard xApps. Furthermore, we test ns-O-RAN with the O-RAN Software Community (OSC) and OpenRAN Gym RICs, simplifying the onboarding from a test environment to production with real telecom hardware controlled without major reconfigurations required. ns-O-RAN is open source and publicly available, together with quick-start tutorials and documentation.
2023
15th Workshop on Network Simulator, WNS3 2023
5G; ns-3; O-RAN; Open RAN; RIC; simulation
04 Pubblicazione in atti di convegno::04b Atto di convegno in volume
ns-O-RAN. Simulating O-RAN 5G systems in ns-3 / Lacava, Andrea; Bordin, Matteo; Polese, Michele; Sivaraj, Rajarajan; Zugno, Tommaso; Cuomo, Francesca; Melodia, Tommaso. - (2023), pp. 35-44. (Intervento presentato al convegno 15th Workshop on Network Simulator, WNS3 2023 tenutosi a Arlington; USA) [10.1145/3592149.3592161].
File allegati a questo prodotto
File Dimensione Formato  
Lacava_Ns-O-RAN_2023.pdf

accesso aperto

Tipologia: Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 1.28 MB
Formato Adobe PDF
1.28 MB Adobe PDF

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/1686006
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 5
  • ???jsp.display-item.citation.isi??? 3
social impact