AI/ML-Driven Optimization for 6G Radio Access Networks
DRIVING-6G (AI/ML-Driven Framework for Joint Dynamic Optimization of Sensing, Computation, Frequency/resource Allocation in 6G RANs) is a collaborative research project labeled under the CELTIC-NEXT framework.
Connected.
Intelligent.
Sustainable.
THE PROJECT
Business Impact & Value Creation
At the PHY/MAC layers, these joint optimizations orchestrate signal processing across transmitter and receiver chains, enhancing system sustainability and computational feasibility to support high throughput and ultra-low latency. Furthermore, integrating sensing techniques with PHY layer optimizations creates a trustworthy, ML-driven resource allocation framework for multi-user environments, unlocking significant business value, reducing operational costs (OPEX/CAPEX) for telecom operators, and enabling novel use cases across autonomous systems, smart manufacturing, and immersive experiences.
Project Duration
Months
July 2026 – December 2028
Partners
Organizations
Leading Universities, Industry & SMEs
Countries
Countries
Across Europe
Objectives
Joint intelligence for real-time 6G optimization
The DRIVING-6G project’s primary objective is to develop and validate an AI/ML-driven framework for join dynamic optimization of real-time sensing, computation offloading, communication, and frequency/resource allocation in 6G Radio Access Networks (RANs).
To achieve its goals, the project leverages cross-layer coordination protocols while decoupling ML algorithms from underlying RAN functions. By integrating AI-powered cognition and collaborative intelligence, DRIVING-6G aims to deliver solutions for multi-objective cross-layer RAN function optimizations in real-time, enhancing system sustainability, supporting high throughput and low latency, and unlocking significant business value for telecom operators and vertical industries.
01
Optimize sensing and communication jointly
Coordinate real-time sensing, communication and frequency/resource allocation through a unified AI/ML-driven framework.
02
Enable cross-layer coordination
Connect PHY and MAC-layer decisions to support multi-objective optimization, high throughput and ultra-low latency.
03
Decouple ML from RAN functions
Create a flexible architecture in which machine-learning algorithms can evolve independently from underlying RAN functions.
04
Deliver collaborative intelligence
Combine AI-powered cognition and shared intelligence to improve sustainability, trust and real-time network performance.
key technicals highlight
Innovations designed for real-world adoption
01. Decoupled ML Framework
Decoupling ML algorithms from underlying RAN functions to create a more flexible and adaptable network architecture.
02. Joint PHY/MAC Layer Optimization
AI/ML-driven orchestration of signal processing across transmitter and receiver chains to support high throughput and low latency.
03. Sensing & PHY Layer Integration
Combining sensing techniques with physical layer optimizations to establish a trustworthy, ML-driven resource allocation framework.
04. Collaborative Intelligence
Intelligence sharing among sensors enhances overall sensing capabilities and supports coordinated decision-making.
05. Multi-Dimensional Business Models
Parallel development and validation of technological innovations and business models can enable customized services, reduced OPEX/CAPEX and new revenue streams.
06. Demo Site Validation
Technological solutions and business models will be tested and validated through Vestel’s demo site to accelerate commercial adoption.
Our consortium
12 partners. 8 countries. One shared vision.
DRIVING-6G brings together telecommunications, research and industrial expertise from across Europe to develop, test and validate the project’s technological and business innovations.
