Samsung, NVIDIA, and Keysight Demonstrate End-to-End AI-RAN Validation Workflow
Case Studies
Samsung Electronics, NVIDIA, and Keysight are advancing a unified validation approach for artificial intelligence radio access network (AI-RAN) development. As artificial intelligence moves deeper into the radio access network, engineering teams need a reliable way to generate realistic datasets, train artificial intelligence / machine learning models, and benchmark performance before field deployment.
Working together, the companies demonstrated an end-to-end AI-RAN validation workflow at Mobile World Congress 2026 in Barcelona. The workflow uses Keysight’s AI-RAN Simulation Toolset with the NVIDIA Aerial Testbed and Samsung-developed AI / ML models to validate AI-driven radio access network behavior in a controlled environment.
For Samsung, NVIDIA, and Keysight, the objective was to bring the AI-RAN validation process into a single automated workflow. That meant connecting data generation, artificial intelligence / machine learning training, and benchmarking so teams can evaluate performance with more consistency and confidence before deployment.
The demonstration showed how a unified test workflow can streamline data generation, artificial intelligence / machine learning training, and performance benchmarking for AI-driven radio access network modules.