Premium
Multiple candidates bit‐flipping algorithms for decoding LDPC codes
Author(s) -
Huang Zhiliang,
Chen Ming,
Diao Chunjuan,
Chen HuaMin
Publication year - 2015
Publication title -
transactions on emerging telecommunications technologies
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.366
H-Index - 47
ISSN - 2161-3915
DOI - 10.1002/ett.2605
Subject(s) - low density parity check code , decoding methods , algorithm , computer science , scheme (mathematics) , set (abstract data type) , bit (key) , berlekamp–welch algorithm , sequential decoding , parity bit , mathematics , error floor , block code , mathematical analysis , computer security , programming language
This letter proposes an effective scheme to improve the error performance of conventional bit‐flipping (BF) algorithms for decoding low‐density parity‐check codes. The proposed scheme, which is referred to as the multiple‐candidates (MC) scheme, constructs a set of candidates for the starting vector of the iterative decoding procedure. Each candidate vector in the set is generated by randomly flipping several high error probability bits of the original hard‐decision vector. Simulation results show that two typical BF algorithms applying the MC scheme achieve noticeable gains over the original BF algorithms with moderate increase in computational cost. Copyright © 2012 John Wiley & Sons, Ltd.