Wireless communication research is wealthy with good ideas and computational power. Yet, there’s a fundamental disconnect between what researchers can simulate and what they will ‌ construct and test. Adjoining fields like machine learning (ML) have flourished with open frameworks and accelerated hardware. But many disruptive ideas never see the sunshine of day as a consequence of the challenges of deployment in cellular infrastructure.
NVIDIA Sionna: Democratizing 6G research
NVIDIA recognized this barrier early and launched NVIDIA Sionna as an open-source library for 6G research, using GPU acceleration. Over 540 scientific publications now reference Sionna, with greater than 200,000 downloads of the source code. Its success stems from openness with comprehensive documentation, textbook-quality tutorials, and trivial installation in Python:
pip install sionnaSionna made rapid prototyping accessible to researchers and developers, even those and not using a GPU. But simulation, nevertheless sophisticated, has limits. You possibly can model channel conditions, but you possibly can’t capture the complete complexity of real-world radio frequency (RF) propagation, including hardware impairments, interference from adjoining cells, or the emergent behaviors of real-world network traffic. To innovate beyond invention, you have to deploy, test, and gather real-world data.
NVIDIA Sionna Research Kit: AI-native 6G lab-in-a-box
The Sionna Research Kit is a real-time, accelerated, fully open platform for wireless research and development. It runs on the NVIDIA DGX Spark and is built on the OpenAirInterface (OAI), providing a whole base station using software-defined radio (SDR) and 5G core network that operates in real-time.
Get your DGX Spark ready, you’re five steps away from running your first simulation:
git clone https://github.com/NVlabs/sionna-rk.git && cd sionna-rk
make prepare-system
sudo reboot
make sionna-rk
./scripts/start_system.sh rfsim_arm64 The Sionna Research Kit isn’t just one other testbed—it’s an open platform that allows you to speed up AI, ML, signal processing algorithms, and ray tracing on a unified memory architecture. No fixed-function accelerators are used. You possibly can inspect the entire telecommunications software stack, modify, and speed up across layers.
From physical layer processing to MAC scheduling to core network routing, your entire system is open for experimentation, including RAN Intelligent Controllers (RIC). Consider it because the wireless equivalent of getting full root access to your infrastructure.
Explore the Sionna Research Kit tutorials
Even the most important projects begin with a single line of code, which regularly is the toughest. The Sionna Research Kit comes with a set of comprehensive tutorials, offering these first lines as blueprints for your individual innovations.
You’ll learn the way the physical layer might be accelerated using GPU-accelerated LDPC decoding. The real-world data acquisition tutorial shows the right way to capture and record real-world 5G signals using the Sionna Research Kit. Next, the Integration of a Neural Demapper tutorial covers training of a neural network-based demapper and integrates it into the 5G stack using NVIDIA TensorRT for real-time inference. Finally, the Software-defined End-to-End 5G Network tutorial allows you to simulate your entire end-to-end system using software-defined user equipment (UE) for the exploration of novel, non-standard-compliant algorithms and protocols.
Digital twin network in real time
Figure 2 shows what you possibly can realize on a single DGX Spark. We deploy a whole base station, but as an alternative of radiating over-the-air (which might require spectrum licenses), we feed the signal right into a GPU-accelerated channel emulation driven by real-time ray tracing. The NVIDIA RT Cores compute physically accurate channel impulse responses in realistic 3D environments. NVIDIA CUDA Cores apply these to the baseband signal while also handling LDPC decoding. As well as, NVIDIA Tensor Cores speed up the PUSCH neural receiver, and its performance is then evaluated.
A business 5G modem connects through cable, experiencing channels physically faithful to real-world RF environments. The whole pipeline uses unified system memory, which avoids unnecessary data movement. An xApp running on a near-real-time RIC monitors live performance metrics as virtual users move through the ray-traced scene. You get a whole interactive digital twin of an RF environment—it’s a 6G lab in a box, with every component of the DGX Spark architecture doing exactly what it’s designed for.
Scaling up: Large-scale radio maps
What you develop on a single DGX Spark is able to scale to the NVIDIA DGX Cloud with the identical code and ray tracing engine, but with orders of magnitude more compute. A single DGX Spark can generate detailed radio maps of a town with a whole lot of base stations in seconds, making real-time network planning possible for local deployments.Â
While you need continental-scale coverage, the cloud takes over. We simulated 5G coverage across the continental US in under five minutes by tracing greater than 35 trillion rays on 96 NVIDIA L40S GPUs (see Figure 3). This can be a fundamental shift in how wireless networks might be planned and optimized, as operators can evaluate latest spectrum allocations, model millimeter-wave deployments in dense urban environments, and integrate non-terrestrial networks (satellite and high-altitude platforms) into existing infrastructure with physics-based accuracy fairly than statistical approximations.Â
Simulating compact environments in real-time on a single DGX Spark and scaling as much as simulate entire countries efficiently redefines what’s possible for deploying the following generation of wireless networks.
Attribution: Google Maps (GEBCO – Landsat / Copernicus, Vexcel Imaging US, Inc., IBCAO, Landsat / Copernicus – Airbus, LDEO-Columbia, NGA, NOAA, NSF, SIO, U.S. Navy), Cesium Ion, OpenStreetMap, Mapzen, OpenCelliD.
Learn more
Visit NVlabs/sionna-rk on GitHub and take a look at the tutorials. The Sionna Research Kit is an element of the NVIDIA AI Aerial portfolio, which incorporates various accelerated computing platforms, software libraries, and tools—enabling developers to construct, train, simulate, and deploy AI-native RAN systems—and move from prototyping to production faster.
