Home5G & BeyondRohde & Schwarz joins AI-RAN Alliance 

    Rohde & Schwarz joins AI-RAN Alliance 


    Company wants to use its test and measurement (T&M) expertise to unlock potential of AI for wireless comms

    Rohde & Schwarz has become the latest member of the recently formed AI-RAN Alliance. The wireless testing company will contribute its T&M expertise to this new collaborative initiative, which aims to integrate artificial intelligence (AI) into wireless communications to advance radio access network (RAN) performance and mobile networks. The company said the membership marks a significant milestone for it on the road to 6G and takes its ongoing collaboration with industry leaders on AI-native air interfaces to a new level. 

    By joining the AI-RAN Alliance, Rohde & Schwarz further aligns itself with key industry players such as Nvidia, Ericsson, Nokia and Samsung and strengthens its role in the development of AI-native air interface test methodologies. Formed in February 2024 during the recent Mobile World Congress in Barcelona, the AI-RAN Alliance will utilise their members’ respective technology expertise and collective leadership to focus on three key areas of research and innovation as the ecosystems moves towards 6G: 

    > AI for RAN: advancing RAN capabilities through AI to improve spectral efficiency. 

    > AI and RAN: integrating AI and RAN processes to utilize infrastructure more effectively and generate new AI-driven revenue opportunities. 

    > AI on RAN: deploying AI services at the network edge through RAN to increase operational efficiency and offer new services to mobile users. 

    Prior to joining the alliance, Rohde & Schwarz has collaborated with Nvidia’s research team on 6G research, pioneering a test bed for exploring neural receiver implementations that promise to revolutionise the air interface by improving performance and network efficiency.  

    A neural receiver constitutes the concept of replacing signal processing blocks for the physical layer of a wireless communications system with trained machine learning models. Academia, leading research institutes and industry experts across the globe anticipate that a future 6G standard will use AI/ML for signal processing tasks, such as channel estimation, channel equalisation, and demapping. 

    Today’s simulations suggest that a neural receiver will increase link-quality and will impact throughput compared to the current high-performance deterministic software algorithms used in 5G NR. 

    To train machine learning models, data sets are an absolute prerequisite. Often, the required access to data sets is limited or simply not available. In the current state of early 6G research, test and measurement equipment provides a viable alternative when generating various data sets with different signal configurations to train machine learning models for signal processing tasks. 

    To run the test bed, the R&S SMW200A vector signal generator emulated two individual users transmitting an 80MHz wide signal in the uplink direction with a MIMO 2×2 signal configuration. Each user was independently faded, and noise is applied to simulate realistic radio channel conditions. The R&S MSR4 multi-purpose satellite receiver acted as the receiver, capturing the signal transmitted at a carrier frequency of 3GHz by using its four phase-coherent receive channels.  

    The data was then provided via the real-time streaming interface to a server. There, the signal was pre-processed using the R&S Server-Based Testing (SBT) framework including R&S VSE vector signal explorer (VSE) micro-services. The VSE signal analysis software synchronized the signal and performed fast Fourier transforms (FFT). This post-FFT data set served as input for a neural receiver implemented using Nvidia Sionna. 

    Nvidia Sionna is a GPU-accelerated open-source library for link-level simulation. It enables rapid prototyping of complex communications system architectures and provides native support to the integration of machine learning in 6G signal processing. 

    As part of the demonstration, the trained neural receiver was compared to the classical concept of a linear minimum mean squared error (LMMSE) receiver architecture, which applies traditional signal processing techniques based on deterministically developed software algorithms. These already high-performance algorithms are widely adopted in current 4G and 5G cellular networks. 

    “Collaboration is critical to unlocking the full potential of 6G technology components, such as an AI-native air interface,” said Rohde & Schwarz CTO Andreas Pauly. “By partnering with leading industry players and joining the AI-RAN Alliance, we ensure that we remain at the forefront of wireless communications innovation. In this way, the ecosystem benefits from our extensive experience in creating holistic test solutions for the wireless industry.”