Listed below are three Keysight launches at MWC which might be price a re-evaluation
On the bottom at MWC the place our workforce was all week, there was one unifying theme throughout keynotes, demonstrations, and hallway conversations: AI.
AI workloads are taking up the community and telcos, knowledge facilities, and enterprises are able to faucet into AI-native options to accommodate the shift.
Naturally, the bulletins that adopted had been one too many. I made a decision to take a divide and conquer strategy as I sift by the hype and discover the most effective MWC check and measurement bulletins to convey to you.
Beginning with Keysight, which, like a number of different T&M suppliers within the house, made a flurry of bulletins on the Congress that, in addition to reflecting the broader AI community push, additionally confirmed effort to deepen its function in AI-centric community deployments.
An unified AI-RAN testing workflow with Samsung, powered by Nvidia
With Nvidia spearheading the AI-RAN (Radio Entry Community) push by the AI-RAN Alliance, discipline trials and checks are underway, paving the trail for early commercialization. Based on Nvidia’s State of AI in Telecommunications report, 66% respondents are investing in or contemplating deploying AI for RAN operations.
Whereas solely 37% mentioned deploying it’s an funding precedence for them, 53% confirmed utilizing AI for reinforcing spectral effectivity within the RAN, and 50% in multi-tenant infrastructure with the purpose to concurrently run AI and RAN workloads.
Workgroups on the AI-RAN Alliance are working in direction of establishing standardized check and validation frameworks for operators and distributors to speed up validation and cut back deployment dangers. To that finish, Keysight and Samsung Electronics collectively developed an AI-RAN testing platform to permit engineers to validate AI-RAN modules with a single workflow.
Crucially, the workflow goals to get rid of the fragmented strategy round knowledge assortment, mannequin coaching, and benchmarking that happen resulting from knowledge coming in from distributed environments, making outcome comparability and validation of PHY-level features particularly difficult.
The answer makes use of Keysight’s AI RAN Simulation Toolset to supply an end-to-end workflow that replaces the fragmented course of with automated knowledge era, AI/ML mannequin coaching, and repeatable benchmarking capabilities. Because of this, engineers can confidently validate RAN conduct utilizing clear insights earlier than discipline testing begins.
The answer, demoed at MWC , is powered by the Nvidia Aerial Testbed for over-the-air AI-RAN analysis, the Nvidia Aerial Omniverse Digital Twin — and AI/ML fashions collectively developed by the three events.
Chief technologist at Keysight, Balaji Raghothaman, mentioned in a press release, “AI within the RAN solely delivers worth when it may be validated with confidence. Working with Nvidia and Samsung, we’re demonstrating a streamlined, automated workflow that unifies knowledge era, AI/ML coaching, and benchmarking, serving to operators and distributors speed up deployment, cut back danger, and make extra knowledgeable selections as they introduce AI-driven RAN capabilities.”
A prototype for AI-driven uplink efficiency with MediaTek
Advancing the RAN work, Keysight collaborated with MediaTek to create a prototype for advancing AI-driven uplink optimization and mannequin life cycle administration for RAN.
Sustaining constant uplink efficiency has turn into a thorny problem for operators as protection circumstances proceed to fluctuate throughout areas. Whereas AI fashions are designed to assist with this, a mannequin educated to work in a single atmosphere fail to ship outcomes based mostly on the discrete circumstances of one other atmosphere. These points weigh heavy on community operations.
The answer, demoed at MWC, seeks to handle this and speed up the transition to AI-native networks. It boosts uplink efficiency by leveraging “RAN-assisted AI decision-making” and evolves AI fashions as required by retraining and over-the-air updates, Keysight mentioned.
VP and GM of Wi-fi Take a look at Group at Keysight, Pen cao mentioned, “Along with MediaTek, we’ve got demonstrated a novel AI-enabling-AI method which helps tackle rising demand on cellular networks uplink efficiency, capability and constant
finish customers’ experiences pushed by AI and AR functions. This entails coaching and testing the
AI-enhanced transmitter mannequin within the UE with high-fidelity real-world coaching datasets generated from Keysight’s RF digital twin channel emulation options with 3D ray tracing capabilities.”
Pre-6G interoperability validation with Ericsson
Final however not the least, Keysight demoed pre-6G troubleshooting and interoperability validation with associate Ericsson. Though not particularly an AI launch, this one is important as AI-nativity is a giant piece of 6G.
With 6G standardization work underway, the answer seeks to supply a full-stack interoperability answer for “pre-6G” whereby operators can validate the early implementations. Some examples Keysight cites are interoperability between prototype base stations and root trigger evaluation use circumstances for a similar.
The answer faucets Keysight’s WaveJudge Wi-fi Analyzer Resolution that analyzes indicators, surfacing threshold deviations and different deep KPI insights, permitting for quicker root trigger evaluation and determination.
“Early validation is vital to advancing next-generation wi-fi. By working
carefully with Ericsson, we’re enabling early exploration of recent concepts alongside the requirements
growth course of. Keysight’s WaveJudge options assist determine points shortly and
speed up our clients’ path from analysis to real-world deployment,” mentioned Lucas Hansen of Keysight.