While 5G has yet to be fully realized — and is only just evolving in the enterprise — the telecom industry has already begun to discuss the forthcoming 6G, which will be even faster, more robust and able to handle increasingly complex network demands.
Still, the next-gen infrastructure will be incredibly difficult to pull off, experts say, pointing to the bombardment of big data and massive network demands.
This will make artificial intelligence (AI) a critical element to 6G — in fact, it will be built in from the very beginning.
“AI will play a critical role in the design and development of 6G, improving network management and optimization,” said Adrian Baschnonga, lead analyst with EY’s global technology, media and entertainment and telecommunications (TMT) group.
5G is living up to expectations — how can we be talking 6G?
Introduced in the late-2010s, 5G is touted for its increased connectivity and bandwidth, higher capacity, lower latency, faster speed and energy efficiency, among other factors.
6G will take this a big step farther, experts say, offering even better performance, speed, low communication delay and wider coverage depth. Still, these advanced capabilities will make 6G networks “immensely complex, requiring more deployment time, cost and management efforts,” write researchers from Nokia Bell-Labs.
Thus, “in the 6G area, AI is becoming more and more important,” write researchers from institute of electrical and electronics engineers (IEEE).
While it’s unclear just when 6G will roll out — it will undoubtedly be an iterative evolution — experts say it will be one of the first AI-native networks, meaning AI will be fully integrated into networking equipment.
“Embedding AI in the network will ultimately reduce the complexity and cost of managing infrastructure for operators, while improving sustainability and security,” said Baschnonga.
This will also spur “more sophisticated” use cases, he said, particularly for industrial customers that rely on predictive and real-time capabilities from data processed and repurposed across a wide swath of device and network endpoints.
AI facilitating 6G networks
AI, by its very nature, constantly mines big data to continuously train and learn. It is also flexible and responsive to network changes. 6G networks will be dealing with explosive data traffic growth and massive device connections, meaning they must have the ability to identify service characteristics and monitor, control and allocate network resources.
Use of AI at the application layer can automate 6G networks and make them more intelligent as they manage untold amounts of big data, researchers assert. AI can also be applied at the physical layer to support large-scale users and antennas and multi-band hybrid transmission.
Many problems in communication systems can’t be modeled, including variation patterns of wireless channels and nonlinear effects of the power amplifier, according to scientists from Vivo Communications Research Institute.
With AI, however, implicit relationships, features and knowledge can be extracted from large amounts of wireless communications data. “Therefore,” Vivo researchers posit “AI capabilities can help to achieve more accurate modeling of complex problems.”
The Nokia Bell-Labs researchers point out that AI can support THz communications at different layers, thus supporting channel acquisition and modulation classification. Similarly, machine learning (ML) at the link layer can benefit beamforming design and channel allocation.
They also note that ML algorithms will be able to improve channel prediction techniques and adapt to real-time changes and localized characteristics, learning specific environments around transceivers.
“6G networks will be more complex due to numerous network topologies, immense growth in the cellular users, staggering data rate demands, complex air interface and vast network coordination methods,” researchers write.
Given this complexity, the derivation of optimum performance is “nearly infeasible” without ML, they contend.
6G will change the human experience
Ultimately, AI-integrated 6G will present untold benefits. In enterprise, Baschnonga said, it will build on existing 4G and 5G and make more efficient use of networking and compute resources. This will in turn lead to speed, coverage and latency gains.
“Taken together,” he said, “this means it will be more suited to data-intensive use cases across different industry sectors, from process automation through to the tactile internet.”
Furthermore, 6G’s advanced frequencies will make it beneficial for multiple use cases in industrial applications, providing “immense data rates,” accelerating IoT and providing wider network coverage, the Nokia Bell-Labs scientists explain.
The “6G-era will bring digital, physical and biological worlds together with the goal to improve human experience and well-being,” they write.
Not to mention, 6G networks will integrate technologies including terrestrial mobile communications, satellite internet and microwave networks, IEEE researchers assert. This will ultimately support group collaboration, intelligent data perception, real-time security assessments and “coordinated coverage of space and earth.”
“Facing the 6G era, the network will usher in new application scenarios and new performance requirements,” they write.
6G will support AI, too
It’s important, however, to view the evolution to 6G as a series of “incremental improvements in mobile,” rather than a simple switch from one standard to another, Baschnonga said.
5G innovation continues in standalone deployments, he said, and that will be compounded by 5G-Advanced. At the same time, early interest in 6G reflects the intensive, multi-year planning cycle for new mobile technologies.
“So, we can expect both technologies to generate discussion side-by-side,” said Baschnonga.
At the same time, 6G and AI will work symbiotically, he and others point out — with 6G technologies supporting increasingly demanding AI use cases.
For instance, 6G will allow AI apps to process nearer the network edge, thus optimizing network services and ensuring more reliable performance and efficiency, said Baschnonga.
“As a result, 6G will play an important role in scaling AI workflows, underlining the symbiotic relationship between these two frontier technologies,” he said.
https://www.sdxcentral.com/articles/feature/how-ai-will-drive-6g-and-why-ai-needs-6g-to-meet-increasing-demands/2024/02/
