University of Georgia researchers are leading the charge on a project that will optimize wireless networks for use with extended reality technologies.
Extended reality is the umbrella term for technologies that immerse users in virtual environments, including virtual, augmented and mixed reality. These applications are extremely demanding on wireless networks, requiring them to respond reliably to user motion, changing environments and strict delay constraints.
The multi-institutional project, Seeing the Channel: Optimizing Perception-aware Environments for XR (SCOPE-XR), brings together researchers from the University of Georgia, Virginia Tech, University of Texas at Austin, University of Wisconsin-Madison and Ericsson. The collaboration spans wireless communications and networking, machine learning, extended reality systems and industry expertise.
The project is part of the U.S. National Science Foundation's recent $110 million investment in improving NextG wireless technology through NSF Verticals-enabling Intelligent Network Systems (NSF VINES). The program aims to enhance U.S. competitiveness in advanced telecommunications technologies.
As the lead institution on the collaborative project, the UGA team is building an experimental open radio access network (O-RAN) that connects devices, such as extended reality headsets, to a wireless network. Its open design allows equipment and software from different suppliers to work together. This platform will allow researchers to move beyond simulation and evaluate the research on a real wireless network. The team includes faculty members Haijian Sun, Zhenlin An, Kyle Johnsen, Mehmet Kurum, WenZhan Song and Jin Sun.
"Our vision is a wireless network that can understand both the physical radio environment and what an extended reality user is experiencing," said Haijian Sun, the project's principal investigator and an assistant professor in UGA's College of Engineering. "By connecting those two kinds of intelligence through O-RAN, we hope to make future immersive applications more responsive, reliable and scalable."
Extended reality headsets require a high-performing wireless network. In real rooms, people move, turn their heads and block signals constantly. To adapt the network for these scenarios, the project will build an AI-native, integrated sensing-and-communication framework, fusing on-device perception (pose, gaze and scene geometry) with radio measurements to predict wireless channel conditions.
Our vision is a wireless network that can understand … what an extended reality user is experiencing.
Haijian Sun, College of Engineering
The long-term goal of the project is to close the loop between the physical environment, extended reality perception, wireless intelligence, network control and the overall extended reality experience.
"We are extremely grateful to the National Science Foundation and the VINES program team for supporting this work and bringing together a broader research community around next-generation wireless networking," Sun said.
The project began in September 2026 and will run through August 2029.
