Keysight and Samsung Demonstrate End-to-End AI-RAN Validation Workflow with NVIDIA

Keysight Technologies, Samsung Electronics, and NVIDIA have collaborated to demonstrate an integrated end-to-end validation workflow for Artificial Intelligence Radio Access Network (AI-RAN) technologies at Mobile World Congress 2026 in Barcelona. This demonstration showcases a streamlined process that integrates Keysight’s AI-RAN Simulation Toolset with NVIDIA’s Aerial Testbed, an over-the-air AI-RAN research platform. Together, the solution automates the validation of AI-driven Radio Access Network modules, specifically highlighting the Physical Uplink Shared Channel (PUSCH) channel estimation use case.

The combined workflow unifies data generation, AI and machine learning training, and performance benchmarking into a single automated system. This integration aims to accelerate innovation cycles and reduce the operational risks of deploying AI-powered RAN capabilities for network operators and vendors. Charlie Zhang, Executive Vice President of Samsung Research America, highlighted that the partnership helps bridge the gap between theoretical AI models and practical, real-world implementation by delivering a comprehensive validation framework. This allows stakeholders to make more informed deployment decisions and enables faster adoption of intelligent radio technologies.

The collaboration represents a significant step in advancing AI applications within telecommunications infrastructure, emphasizing the importance of automated validation in ensuring reliable and efficient AI-RAN rollout. It also reflects the broader industry trend towards embedding AI in network management to improve performance, capacity, and adaptability.