Forward Backward Syndrome Computation: A Reduced Complexity CRC Code Decoder

Titouan Gendron (1, 2) , Emmanuel Boutillon (1) , Charbel Abdel Nour (1, 3) , David Gnaedig (2)

1 Lab-STICC_CODES - Equipe CODES
2 TurboConcept
3 IMT Atlantique - MEE - Département Mathematical and Electrical Engineering

This paper proposes a simplified method for correcting a single-bit error event in a message protected by a CRC code thanks to a new decoding method called Forward-Backward Syndrome Computation. The two key ideas of the paper are to show that: 1) syndrome computation can be performed in any direction (Forward and Backward), 2) consequently, it is possible to create a partition of the received message in a set of disjoint syndromes, thus allowing a direct identification of the error position. Evaluation in terms of number of 2-input binary logic functions shows a complexity reduction by a factor of up to 3 compared to conventional methods.

Mots clés: CRC code; error correction; single-bit error; syndrome computation

Domaines: Réseaux et télécommunications [cs.NI]

Titouan Gendron, Emmanuel Boutillon, Charbel Abdel Nour, David Gnaedig. Forward Backward Syndrome Computation: A Reduced Complexity CRC Code Decoder. IEEE Communications Letters, 2023, 27 (5), pp.1267 - 1271. ⟨10.1109/LCOMM.2023.3264162⟩. ⟨hal-04073460⟩