The adoption of the Internet of Things (IoT) for digital transformation is sketching new forms of distributed data and decentralised machine learning, localised with data sources, or in a proximity to adhere to such requirements as data privacy and possibly latency. The new form of data and machine learning is opening new horizons in future networks (such as 6G) while introducing novel requirements that leverage the role of data semantics at the network layer.
In this talk, I will share our vision on how future networks can find and exchange semantically-relevant data for such applications as machine-learning. Specifically, I will show how state-of-the art information-centric networks can be extended with novel networking services to enable semantic-based mapping and delivery of context-relevant, not necessarily exact copies, of data.
The talk is based on a recent publication: “Flexible Semantic-based Data Networking for IoT Domains” in IEEE HPSR 2021.