AbstractSDRs: Bring down the two-language barrier with Julia Language for efficient SDR prototyping

Corentin Lavaud (1) , Robin Gerzaguet (1) , Matthieu Gautier (1) , Olivier Berder (1)

1 GRANIT - Algorithmes et architectures adaptatifs pour les systèmes sans-fils efficaces en énergie

This paper proposes a new methodology based on the recently proposed Julia language for efficient Software Defined Radio (SDR) prototyping. SDRs are immensely popular as they allow to have a flexible approach for sounding, monitoring or processing radio signals through the use of generic analog components and lot of digital signal processing. As, in this paradigm, most of the processing is done at software levels (i.e. on a CPU), an efficient software methodology has to be envisioned. Right now, most of the existing methods focus on lowlevel languages (C or C++) for good runtime performance (at the cost of easy prototyping) or high-level language (such as Python) for flexibility (at the price of runtime performance). In this article we propose a new methodology based on Julia language that addresses this two-language problem and paves the way for efficient prototyping without giving up runtime performance. To prove the benefits of the proposed approach, a performance benchmark with several optimisation levels compares the Julia approach with C++ and Python.

Mots clés: Software Defined Radio; Julia language; fast prototyping; benchmark

Domaines: Langage de programmation [cs.PL]; Systèmes embarqués; Traitement du signal et de l'image [eess.SP]

Corentin Lavaud, Robin Gerzaguet, Matthieu Gautier, Olivier Berder. AbstractSDRs: Bring down the two-language barrier with Julia Language for efficient SDR prototyping. IEEE Embedded Systems Letters, 2021, 13 (4), pp.166-169. ⟨10.1109/LES.2021.3054174⟩. ⟨hal-03122623⟩