NVIDIA is making it easier for everyone to help build the next generation of mobile networks by open sourcing its Aerial software. This means that anyone can access, use and improve this software without limits. Developers will be able to use it on various NVIDIA platforms, including the powerful and compact DGX Spark supercomputer, to create and test new ideas quickly. What used to take months or years can now be done in days.
With open source Aerial, people all over the world can work together on AI-powered 5G and 6G networks. Unlike old systems that were closed off and used exclusive hardware and software, this open approach encourages fast innovation and teamwork. The software will be available on GitHub, making it easy for researchers and developers everywhere to access.
Aerial software comes with many helpful features. It has a framework that turns Python code into faster CUDA code that runs on NVIDIA computers. It also includes AI models that can make wireless networks smarter and tools that let developers build their own custom network solutions. Developers can even create new apps that interact with network data in real time. These features have already made it possible to build the first fully American-made AI-powered wireless technology, which is paving the way for future 6G applications.
Opening up Aerial to everyone does not just help phone companies—it also lets experts from other fields create new kinds of mobile apps. These could be advanced AI programs that need very reliable and fast connections. By sharing its tools so widely, NVIDIA is helping the United States lead in open source technology and changing how new ideas in telecom are developed.
NVIDIA’s DGX Spark is a small but powerful computer designed for AI research in wireless networks. Using DGX Spark, researchers can set up and test their own 5G and 6G networks at a desk. With kits like Sionna Research Kit, you can even set up a live 5G network in an afternoon. This portability means experiments do not have to stay in a lab, making research faster and more flexible.
The Aerial Testbed, now working with DGX Spark, allows developers to test out their full 5G and 6G networks with real data and hardware. This helps speed up the process of bringing new products to market. Companies like Dell are joining in too, with products built on NVIDIA technology to help tackle the challenges of 6G.
Thousands of researchers and organizations worldwide are already using NVIDIA’s AI Aerial tools, from major universities like MIT to innovative startups. Industry groups like the AI-RAN Alliance are using these tools to help shape future mobile networks. With open access and a focus on sharing, NVIDIA is starting a new chapter in wireless innovation by inviting anyone interested to contribute and push the boundaries of what is possible in an AI-driven world.
Original article and image: https://blogs.nvidia.com/blog/open-source-aerial-ai-native-6g/
