The wireless industry anticipates that a future 6G standard supports an Artificial Intelligence / Machine Learning (AI/ML) based air interface natively. In this joint demonstration, Rohde & Schwarz and Nvidia demonstrate a neural receiver approach, using a trained machine learning model for signal processing tasks such as channel estimation, channel equalization, and demapping. Due to the lack of a 6G standard, we showcase a 5G NR PUSCH multi-user MIMO scenario, emulating two users with 2x2 MIMO, that are independently faded to simulate realistic channel conditions using the R&S SMW200A vector signal generator. The signal is captured using the MSR4 satellite receiver and transferred to a server. Our server-based testing (SBT) framework, including vector signal explorer (VSE) microservices is used to pre-process the data. The generated post-FFT data is input to Nvidia's SIONNA software framework, an open-source library for 6G physical layer research. We showcase the performance based on a BLER over SINR measurement and compare it to ideal, perfect channel knowledge performance and traditional receiver implementation used in today's 4G LTE and 5G NR networks.
