Researchers Propose Blueprint to Fuse Wireless Technologies and AI
Published in the Proceedings of the IEEE journal, a paper by Virginia Tech researchers suggests that the next significant leap in wireless technology hinges on the integration of artificial general intelligence (AGI), an AI capable of human-like thought, imagination, and planning. This shift, the researchers argue, is essential for creating truly revolutionary advancements in wireless networks.
According to Walid Saad, professor in the College of Engineering and the Next-G Wireless Lead at the Virginia Tech Innovation Campus, the current limitations in AI, particularly a lack of common sense, prevent it from achieving the full potential of wireless technology. “We’re looking at least 10 or 15 years down the line before we have a wireless network with artificial general intelligence [AGI] that can think, plan, and imagine,” Saad stated. The research team consists of Saad, Ph.D. student Omar Hashash, and postdoctoral associate Christo Thomas.
The Missing Link: AGI in Wireless
The core argument of the paper is that previous generations of wireless networks, even the anticipated 5G to 6G upgrade, which incorporates AI-driven elements, have not gone far enough. The researchers contend that what is missing is the ability for wireless networks to use and respond to information using something akin to common sense. Current AI systems excel at extracting patterns from data but fall short in novel or unexpected scenarios, a capability humans master with ease.
“The next generation of wireless networks and AI are converging hand in hand, but few are seeing how they can actually be seamlessly merged,” Hashash said.
“Common sense allows us to deal with new scenarios, learn by analogy, and connect the dots to fill in missing plausible elements when needed,” – Walid Saad
Blending Physical and Virtual Worlds
The researchers focused their work on the metaverse and the use of digital twins. These twins are digital replicas of real-world entities and environments. They believe the metaverse, with its digital twins, provides a crucial opportunity for networks to develop perception, world models, planning abilities, and logical reasoning.
The development of digital twins can give wireless networks the perception needed to allow for the seamless integration of physical, virtual, and digital dimensions, something that only a network with human-level thinking can achieve. This integration would facilitate fully immersive experiences, such as using a VR headset to travel across space and time from one’s home.
The “Telecom Brain” Blueprint
To achieve this vision, the researchers propose a fundamental shift beyond the AI-native wireless systems expected with 6G. Their model calls for networks that possess human-level intelligence, achieved through experiences at the intersection of the digital world and future wireless networks. This AGI-native network could then bring common-sense abilities to digital twins, enabling a new generation of human-like AI entities.
“We must create wireless networks with intrinsic abilities to understand the mathematical mechanics behind the designed AI models, the physical properties of real-world objects, and their interactions with each other,” said Thomas. “This requires us to fuse mathematical principles, category theory, and neuroscience to model the physical world and understand the complex operations of the human brain.”
This paradigm shift proposes that the missing link lies with the wireless network and its digital twin components, like using an existing twin as a world model to enable human-level thinking and incorporating these processes into the wireless network.
