How NVIDIA Open Source Tools Are Helping Researchers Build Next-Generation 6G Networks

Over the past year, NVIDIA's open source 5G and 6G tools, such as Aerial and Sionna, have surpassed one million downloads and are used by over 8,000 research and academic institutions globally. These tools connect simulation and AI training with real-world wireless experiments, helping researchers tackle key challenges in future networks. At King's College London, researchers used NVIDIA Sionna to develop goal-oriented semantic communications for robots. Instead of sending full images, robots transmit compact object descriptions, reducing task completion time by up to 52.6% in dynamic environments. Meanwhile, SoftBank and ETH Zurich demonstrated that AI-powered neural receivers can significantly improve signal reception and throughput under interference. Energy management and simulation speed are also major areas of improvement. Researchers at the University of Leeds used a portable 5G setup powered by NVIDIA to predict network demand, cutting base station power consumption by 45.6%. Additionally, the Centre Tecnològic de Telecomunicacions de Catalunya used GPU acceleration to reduce network simulation runtimes from hours to minutes, speeding up the path from innovation to real-world deployment.