NVIDIA Reports 10x Growth for Open Source 6G Tools
NVIDIA says global developer adoption of its AI Aerial and Sionna software has grown 10x over the past year, while its open source 5G and 6G tools have passed 1 million downloads. More than 8,000 research and academic institutions worldwide are using the platforms to test network designs and explore AI-native 6G applications.
- NVIDIA says adoption of AI Aerial and Sionna grew 10x over the past year.
- Its open source 5G and 6G tools have surpassed 1 million downloads.
- More than 8,000 research and academic institutions use the platforms.
- Projects address robotic response times, signal reception and network power use.
- A simulator integration reduced runtime from approximately three hours to 10-20 minutes.
An open source foundation for 6G research
NVIDIA said its Aerial wireless software joined Sionna and a broader group of open source tools after the company released Aerial as open source a year ago. The tools are designed to connect wireless simulation and AI training with experiments on real over-the-air systems.
According to NVIDIA, this gives researchers a shared foundation for testing ideas, comparing results and building on one another’s work. The company said the projects now emerging point toward networks that combine connectivity, computing and sensing on a converged, software-defined platform for billions of people, machines and agents.
The tools are being used to explore networks that can respond more effectively to signals, keep robots and autonomous systems connected, adapt as devices move and manage energy use as demand changes.
Projects target robots, reception and power use
At King’s College London, researchers connected a camera-equipped robot arm to an NVIDIA Jetson AGX Orin running NVIDIA Sionna. Their goal-oriented semantic communications approach sent compact descriptions of objects, positions and relationships instead of periodic full images.
In fruit-packing and cube-stacking experiments with deliberately introduced faults, the approach reduced robot task completion time by up to 52.6% and improved task success, according to NVIDIA. The system also detected dropped objects, recovered from unsuccessful actions and, in configurations using a digital twin, checked planned movements for possible collisions before execution.
SoftBank integrated a neural receiver into NVIDIA Aerial CUDA-Accelerated RAN, or ACAR, and tested it with a commercial 5G smartphone. In controlled interference tests, the receiver produced higher uplink throughput than a conventional receiver at all 15 tested settings, with a maximum gain of roughly 60%, NVIDIA said.
At the University of Leeds, researchers developed a system that predicts network demand 10 minutes ahead and uses the forecasts to guide base station sleep decisions. The team reported a 45.6% reduction in RAN power consumption relative to a full-power baseline while maintaining device connectivity.
Faster simulation could speed network design
ETH Zurich researchers used local measurements to tune neural receivers and found that the approach reduced block error rates by more than one-third compared with selected pretrained versions, according to NVIDIA.
Separately, the Centre Tecnològic de Telecomunicacions de Catalunya brought GPU-accelerated radio channel models running on ACAR into its open source 5G-LENA network simulator. In an early evaluation, the integration reduced simulation runtime from approximately three hours to 10-20 minutes, depending on the integration, while preserving expected statistical behavior and network performance.
That reduction could allow researchers to compare more designs, investigate more demanding scenarios and refine network concepts before deploying equipment or conducting field trials. NVIDIA AI Aerial and NVIDIA Sionna are available on GitHub, and researchers and developers will showcase further work in 6G Developer Sessions at NVIDIA GTC Washington, D.C.
