A key concept addressed in the 6G-LEADER Project is the optimization of communication processes by aligning the information exchange with specific objectives. This communication paradigm, commonly referred as Goal-Oriented Semantic Aware Communication, goes beyond the legacy and well-known Shannon’s communication paradigm focusing on the reliable transmission of bits. Instead, it emphasises the information relevance with respect to the exchange intent. To this end, the 6G-LEADER project investigates automated frameworks for extracting information meaning, allowing networks and services to exchange only the most relevant data, thus enhancing 6G networks sustainability.
What does this technology consist of?
The Goal-Oriented Semantic-Aware Communication paradigm seeks to enhance traditional communication systems with the capability to extract goal-driven latent meanings embedded in transmitted information. By doing so, network services aligned with these specific goals can adopt optimized operational policies. Artificial intelligence (AI), particularly deep learning techniques, plays a key role in this process. AI frameworks can effectively construct and refine goal-oriented information latent representations while simultaneously providing network policies for RAN performance optimization according to the intended objectives.
Why is it relevant for 6G networks?
moving beyond conventional data transmission toward extracting and conveying only the most relevant semantic information behind the large amount of raw data. This approach reduces radio access network operational costs without compromising the ability to anticipate and optimize the use of key resources, such as radio and computing. Moreover, network services can dynamically adapt radio access network behaviour by relying on semantic data that is relevant to the specific goals pursued.
How do we approach this in the 6G-LEADER project?
The 6G-Leader project envisions integrating semantic information extraction capabilities within the Open RAN domain. By leveraging advanced AI/ML techniques, the 6G-Leader framework can generate semantic latent representations that capture the most relevant features of the network state. These embeddings only bring relevant knowledge, allowing the transmission of the information necessary for decision-making and control, only.
Building on this foundation, the 6G-Leader architecture onboards semantic-aware network services that enable the network to learn and optimise itself. These AI/ML-based services continuously monitor the network through the semantic embeddings and provide control actions aimed at achieving specific goals such as maintaining an optimal trade-off between energy efficiency and quality of service.
What are the benefits?
The capability of networks to become semantic-aware, thereby enabling more efficient decision-making, brings significant benefits for telecommunications operators. Indeed, more effective management of network resources can be achieved, as the network can autonomously evaluate its operational status and execute appropriate actions within very short timeframes, eliminating the need for human intervention in both the assessment and decision-making processes. Furthermore, the application of such technology to energy efficiency brings notable societal benefits, as it directly supports several of the United Nations Sustainable Development Goals (SDGs) related to sustainable infrastructure, and environmental responsibility.
