The next global mobile communication standard 6G strives to push the technological limits of radio frequency (RF) communication even further than its predecessors: Data rates beyond 100 Gbit/s, RF bandwidths above 1 GHz, and sub-millisecond latency necessitate very high performance development tools to enable the extent of innovation required for 6G's likely features. We propose a new SDR firmware and software architecture designed explicitly to meet these challenging requirements. It relies on Ethernet and commercial off-the-shelf network and server components to maximize flexibility and to reduce costs. We analyze state-of-the-art solutions (USRP X440 and other RFSoC-based systems), derive architectural design goals, explain resulting design decision in detail, and exemplify our architecture's implementation on the XCZU48DR RFSoC. Finally, we prove its performance via measurements and outline how the architecture surpasses the state-of-the-art with respect to sustained RF recording while maintaining high Ethernet bandwidth efficiency. Building a micro-Doppler radar example, we demonstrate its real-time and rapid application development capabilities.
In this paper, we implement a low-latency rapid-prototyping platform for signal processing based on software-defined radios (SDRs) and off-the-shelf PC hardware. This platform allows to evaluate a wide variety of algorithms in real-time environments, supporting new developments in the fields of classical, AI-based, and hybrid signal processing. To accomplish this, the streaming protocol of the used USRP X310 devices is implemented using the Data Plane Development Kit (DPDK), which allows to handle network communication in userspace only. This bypasses the kernel and thus avoids the latencies caused by interrupt handling, scheduling, and context switches. It allows signal processing to be performed on isolated processor cores that are protected from interrupts to a great extent. To validate our approach, linear time-invariant channel emulation has been implemented. For this, an analog-to-analog latency of 31 microseconds was achieved, demonstrating that our PC-based approach enables the implementation of rapid-prototyping systems with low latency.