6G’s Next Evolution: AI, Spectrum Efficiency, and Smarter Design
As the wireless industry looks ahead to 6G, conversations are shifting from speculation to strategy. In a recent interview with Javier Campos, 6G Physical Layer Architect and Technical Lead at Keysight, I explored the technical and architectural directions that are beginning to take shape. From deeper artificial intelligence (AI) integration to smarter spectrum use and sensing capabilities, the insights shared reflect a growing focus on purposeful innovation, building on what works, refining what doesn’t, and designing with efficiency in mind.
AI: Deep Integration, Real Impact
AI and machine learning are poised to play a central role in 6G, but not just as buzzwords. Campos emphasized that the real value lies in embedding intelligence into the system from day one:
“AI and machine learning won’t just be layered on top, they’ll be deeply woven into the fabric of 6G from day one.”
He sees AI enhancing physical layer functions like channel state information (CSI) feedback and beam management, but cautions that the industry must still grapple with power consumption and training complexity:
“We need to understand the full trade-offs especially around power consumption and training complexity. Without that clarity, we risk adding overhead without meaningful gains.”
Spectrum: Smarter, Not Just More
While frequency range 3 (FR3) is gaining traction, Campos highlighted a shift in mindset: while chasing new bands, operators are looking to leverage existing 5G infrastructure.
“There’s a strong push to reuse deployed 5G hardware for 6G, which introduces design constraints but also forces us to be more resourceful.”
Efficiency is key, not just in spectrum usage, but in how we aggregate carriers and manage non-contiguous bands:
“It’s not just about accessing new spectrum like FR3. It’s about building systems that operate efficiently across non-contiguous bands and make smarter use of what’s already available.”
Rethinking the Stack: From Procedures to Use Cases
Campos predicts that 2026 will be a pivotal year for 6G research and development (R&D), with three areas leading the charge:
- Uplink coverage: There will be a lot of research in how to enhance uplink waveforms.
- AI/ML in the PHY: The deeper integration of AI and machine learning in the physical layer opens up new use cases.
- Simplified procedures: There will be a lot of rethinking about the physical layer and higher layer procedures, all of the procedural aspects of the specification will be reconsidered, simplified and enhanced.
Joint Communication and Sensing: Built-In from the Start
Sensing isn’t just an add-on, it’s a foundational element of 6G. Campos sees unmanned aerial vehicles (UAVs) as the first use case, but the design must be flexible enough to support others.
“We’re exploring whether communication waveforms can serve both communication and sensing, and if not, design new ones for sensing that will coexist with those used for communication purposes.”
Debunking the “Revolution” Narrative
Despite the excitement around 6G, Campos offers a grounded perspective:
“There’s a common misconception that 6G will be a complete overhaul. In reality, it will be revolutionary in some areas but largely be an evolution and refinement of what we’ve built in 5G.”
He also notes that while sub-terahertz (sub-THz) frequencies once dominated the conversation, they’ve now taken a backseat and are not currently in scope for 3GPP’s 6G release.
“Sub-THz was a hot topic early on, but it’s no longer a priority for the initial 3GPP 6G release. The focus has shifted to more practical, near-term spectrum strategies.”
Conclusion: Design with Purpose
Campos’ insights reinforce a central theme emerging in 6G: innovation must be purposeful, measurable, and aligned with real-world constraints. Whether it’s AI, spectrum, or sensing, the future of 6G lies in designing systems that deliver tangible value without overengineering.”
Listen to the full interview here.