BROWSE
Detail
Conference Paper
Data Acquisition and Visualization for AI/ML-based Radio Resource Management Optimization in the ns-O-RAN Framework
Cited 0 time in 
Share 
Share
- Authors
- Seung-Eun Hong, Jungmo Moon, Jaewook Lee
- Issue Date
- 2024-07
- Citation
- International Conference on Ubiquitous and Future Networks (ICUFN) 2024, pp.476-478
- Publisher
- IEEE
- Language
- English
- Type
- Conference Paper
- Abstract
- The transformative shift from conventional Radio Access Network (RAN) architectures to the innovative Open RAN (O-RAN) has sparked considerable interest due to its visionary principles of open interfaces, virtualization and cloudification, and AI/ML-driven automation. Central to the advancement of AI/ML-based intelligent radio resource management in O-RAN is the acquisition, management, and visualization of key performance measurement (KPM) data across diverse RAN environments. This paper extends the ns-O- RAN framework by leveraging NS-3 simulation to generate comprehensive cell- and UE-level data, subsequently converted into a software-defined real-world network interfacing with the E2 interface. The periodic transmission of RAN data to the RAN intelligent controller facilitates efficient data separation into cell and UE measurements, stored in InfluxDB for optimized time series data management. Furthermore, we employ Grafana in tandem with InfluxDB to develop a user-friendly data visualization dashboard, enabling to effortlessly access and interpret a wide array of performance metrics.
- KSP Keywords
- Data Acquisition(DAQ), Data Acquisition and Visualization, Data separation, Key Performance, Performance measurement, Radio Access Network(RAN), Radio Resource Management(RRM), Real-world, Software-Defined, Time series data management, User-friendly