Software-Defined Wireless Communication Systems for Heterogeneous Architectures
Open Access
- 2 October 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Abstract
Future cellular networks will be programmable and increasingly software-defined with APIs to hook into the communication stack closer and closer to the physical layer. This allows operators, for example, to plug-in third-party machine learning algorithms to optimize performance. At the same time, this flexibility implies that compute resources cannot be provisioned statically but have to be distributed dynamically during runtime. While the hardware platforms for such systems are available, we lack suitable software frameworks that help to realize such systems. In this demonstration, we present two Open Source projects that fill this gap: FutureSDR, a portable real-time signal processing framework with native support for accelerators (like GPUs and FGPAs); and IPEC, which enables fully automatic composition of multi-accelerator FPGA designs. We believe that these tools - especially in combination with each other - can provide the base for building research prototypes and allow experimentation with software-defined wireless communication systems.Keywords
Funding Information
- Deutsche Forschungsgemeinschaft (SFB 1053 MAKI)
- Bundesministerium für Bildung und Forschung (16KISK014)
- LOEWE initiative (emergenCITY)
This publication has 4 references indexed in Scilit:
- Understanding O-RAN: Architecture, Interfaces, Algorithms, Security, and Research ChallengesIEEE Communications Surveys & Tutorials, 2023
- IPEC: Open-Source Design Automation for Inter-Processing Element Communication2022
- dApps: Distributed Applications for Real-Time Inference and Control in O-RANIEEE Communications Magazine, 2022
- On the GNU Radio Ecosystem2015