KIT | KIT-Bibliothek | Impressum | Datenschutz

Spiking Neural Belief Propagation Decoder for Short Block Length LDPC Codes

von Bank, Alexander 1; Edelmann, Eike-Manuel 1; Miao, Sisi 1; Mandelbaum, Jonathan 1; Schmalen, Laurent 2
1 Karlsruher Institut für Technologie (KIT)
2 Communications Engineering Lab (CEL), Karlsruher Institut für Technologie (KIT)

Abstract:

Spiking neural networks (SNNs) are neural
networks that enable energy-efficient signal processing due to
their event-based nature. This letter proposes a novel decod-
ing algorithm for low-density parity-check (LDPC) codes that
integrates SNNs into belief propagation (BP) decoding by approx-
imating the check node update equations using SNNs. For the
(273,191) and (1023,781) finite-geometry LDPC code, the pro-
posed decoder outperforms sum-product decoder at high signal-
to-noise ratios (SNRs). The decoder achieves a similar bit error
rate to normalized sum-product decoding with successive relax-
ation. Furthermore, the novel decoding operates without requir-
ing knowledge of the SNR, making it robust to SNR mismatch.

Zugehörige Institution(en) am KIT Communications Engineering Lab (CEL)
Publikationstyp Zeitschriftenaufsatz
Publikationsmonat/-jahr 01.2025
Sprache Englisch
Identifikator ISSN: 1089-7798, 1558-2558, 2373-7891
KITopen-ID: 1000178423
Erschienen in IEEE Communications Letters
Verlag Institute of Electrical and Electronics Engineers (IEEE)
Band 29
Heft 1
Seiten 45–49
Nachgewiesen in Scopus
Web of Science
OpenAlex
Dimensions

Verlagsausgabe §
DOI: 10.5445/IR/1000178423
Veröffentlicht am 24.01.2025
Originalveröffentlichung
DOI: 10.1109/LCOMM.2024.3492711
Scopus
Zitationen: 2
Web of Science
Zitationen: 1
Dimensions
Zitationen: 3
Seitenaufrufe: 32
seit 24.01.2025
Downloads: 16
seit 29.01.2025
Cover der Publikation
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page