VIAVI unveils 6G city-scale digital twin for AI training

Viavi Solutions Inc. (VIAVI) shared milestones in 6G and AI research based on the company’s 6G Forward Program. As 6G is the first AI-native generation of wireless communications, data sources such as a 6G city-scale digital twin will be vital for training other elements of the network such as the 6G AI-Native Air Interface.

6G City-Scale Digital Twin

In 2022, VIAVI announced support for research at the Institute for Wireless Internet of Things and the Open6G cooperative research center at Northeastern University, as part of the VIAVI 6G Forward Program. A key aspect of the research was large-scale RF propagation channel modeling based on AI and ML technologies to develop a city-scale digital twin of a 6G network.

The digital twin research has expanded to include:

Use of AI/ML to augment ray tracing for radio frequency (RF) propagation modeling to drive the digital twin
Conducting real-world measurement campaigns in the Northeastern Boston campus using the VIAVI Ranger rugged RF capture and playback instrument
Incorporating higher-layer KPIs in the digital twin, to more effectively model effects of network conditions on application performance.

“Our research collaboration with Northeastern University has led to breakthroughs in the use of AI/ML for radio propagation modeling,” said Dr. Ian Wong, CTO Office, VIAVI. “We believe this research is critical in enabling large-scale network digital twins, which VIAVI is well poised to deliver given our leading network emulation solutions in the lab and network monitoring and optimization solutions in the field, gathering data from actual operator networks.”

“At Northeastern's Institute for the Wireless Internet of Things (WIoT), we have been using VIAVI's state-of-the-art testing and measurement equipment to model and deploy high-fidelity digital twins of real-world wireless networks on our Open RAN Digital Twinning platform, Colosseum,” said Professor Tommaso Melodia, Director of WIoT at Northeastern University. “This will facilitate the research, development, and experimentation of AI-driven solutions for 5G and beyond networks in a risk-free digital-twinned world with hardware in the loop.”

6G AI-Native Air Interface (AI-AI)

The AI-Native Air Interface (AI-AI) has appeared as a shift in the conventional methods of developing, standardizing, and commercializing communication technologies with the goal of effectively supplying data to applications while addressing communication problems and hardware limitations. In the physical layer of wireless communications, the neural receiver concept suggests replacing traditional signal processing blocks with trained AI models.

To enable comprehensive testing of neural receivers, VIAVI has developed and demonstrated its own receiver as part of an end-to-end test and training environment. The base station employs an AI-aided constellation design, where the modulation block is replaced with a neural network to design custom constellation for the receiver.

The VIAVI in-house developed neural receiver is generalizable over various channel models and modulations, delay and doppler, signal-to-noise ratio (SNR) and other conditions, which is vital for deployment in practical systems. VIAVI used open-source and in-house libraries to implement the transceiver architecture.

“VIAVI continues to explore and develop groundbreaking future generations of wireless technology, and we are at the frontier of AI-native 6G,” said Dr. Stephen Wang, Managing Director, VIAVI Marconi Labs. “From neural receivers to rate-splitting multiple access for semantic networks, our 6G research helps the industry create networks that dynamically adapt to different environments and conditions. We look forward to working with research labs and network equipment manufacturers to accelerate development of these technologies.”