From the source
NVIDIA says global developer adoption of its AI Aerial and Sionna software has grown 10x in the year since it open sourced Aerial, with its open source 5G/6G tools passing 1 million downloads.
- More than 8,000 research and academic institutions now use the platforms, which connect simulation and AI training with experiments on real over-the-air wireless systems.
- King's College London connected a camera-equipped robot arm to an NVIDIA Jetson AGX Orin running Sionna, and goal-oriented semantic communications cut robot task completion time by up to 52.6% in fruit-packing and cube-stacking tests with deliberately introduced faults.
- SoftBank integrated a neural receiver into NVIDIA Aerial CUDA-Accelerated RAN and, in controlled tests with interference using a commercial 5G smartphone, it delivered higher uplink throughput than a conventional receiver at all 15 tested settings, with a maximum gain of roughly 60%.
- ETH Zurich found that tuning neural receivers with local measurements reduced block error rates by more than one-third compared with selected pretrained versions.
- The University of Leeds reported a 45.6% reduction in RAN power consumption against a full-power baseline using demand forecasts 10 minutes ahead to guide base station sleep decisions.
- CTTC brought GPU-accelerated radio channel models running on ACAR into its open source 5G-LENA simulator, cutting an early evaluation's simulation runtime from about three hours to 10-20 minutes.






