The International Telecommunication Union organised the ITU Workshop on Machine Learning for 5G and beyond which took place on 17 June 2019 in ITU headquarters, in Geneva
Dr. Henna presents at ITU Meeting on ‘Machine Learning for Future Networks including 5G’ in Geneva. The International Telecommunication Union (ITU) organized the ITU Workshop on Machine Learning for 5G and beyond which took place on 17 June 2019 in ITU headquarters, in Geneva, Switzerland.
The workshop was followed by the 6 meeting of the Focus Group on Machine Learning for Future Networks including 5G (FG-ML5G), from 18 to 20 June 2019, at the ITU premises.
Postdoctoral Researcher, Shagufta Henna, is the author of two documents entitled ‘Requirements, architecture and design for machine learning function orchestrator (MLFO)’ and ‘Energy Efficient Trusted Multi-tenancy in 5G-Crosshaul’ both of which were discussed during the ITU meeting last week. Shagufta’s research will play an important role in into an upcoming ITU standard on machine learning for 5G.
Discussions
As described in “Draft Recommendation ITU-T Y.3172 (formerly Y. IMT2020-ML-Arch): Architectural framework for machine learning in future networks including IMT-2020”, the MLFO plays an important role in managing machine learning models in the operator’s network. Dr. Henna defines the requirements, architecture and design of the MLFO which will one of the future ITU recommendations.
The following are the most important functionalities of the MLFO:
- MLFO is a logical orchestrator that can monitor and manage the nodes in a machine learning (ML) pipeline. It supports the deployment of ML pipeline in a variety of ML underlay networks.
- Optimal placement of ML pipeline nodes is achieved in the network with the help of MLFO, coordinated with management plane of the ML underlay networks.
- MLFO lets the network operator specify the ML application using a declarative specification, abstracted from the ML underlay networks. This helps in standard representation, handling and comparison of ML applications in the operator’s network. This also helps in clear separation of technology-specific interfaces and the ML application. Such a specification is important for interoperable integration of ML marketplaces “Architecture for ML marketplace integration in future networks including IMT-2020 (China Unicom)” to operator’s network.
- Chaining and split of ML pipeline nodes, selection of ML models, monitoring their performance, reselection and update, if needed, are achieved using MLFO.
TU-T Focus Group
The International Telecommunication Union (ITU) Focus Group on Machine Learning for Future Networks including 5G was established by ITU-T Study Group 13 at its meeting in Geneva in 2017. The Focus Group will draft technical reports and specifications for machine learning (ML) for future networks, including interfaces, network architectures, protocols, algorithms and data formats.
Participation at the Focus Group was open to ITU Member States, Sector Members, Associates and Academic Institutions and to any individual from a country that is a member of ITU. This included individuals who are also members of international, regional and national organizations.