… In this paper, we proposed a new design of linear finite alphabet iterative decoding rules
of LDPC codes over BSC via recurrent quantized neural networks. The RQNN framework can …

… NS) finite alphabet iterative decoders (FAIDs) for quasi-cyclic … We employ recurrent quantized
neural networks to optimize … To cope with this, many coarsely quantized LDPC decoders […

… the finite precision effect, we propose a new neural network-… scaling and quantization to each
node for irregular LDPC codes. In … -inference precoding for coarsely quantized massive MU-…

… Lin, “Finite alphabet iterative decoding of LDPC codes with coarsely quantized neural networks,”
in 2019 … Lin, “Designing finite alphabet iterative decoders of LDPC codes via recurrent …

… LDPC codes [3]. The wide range of node degrees and required support for rate compatibility
make the decoder design for very coarse quantization … the neural network finite alphabet (…

… The LDPC codes and optimized weights in this paper can be found and downloaded in [18]. …
Schmalen, and AG i Amat, “Pruning and quantizing neural belief propagation decoders,” …

… ) LDPC code show that a layered Min SumRCQ decoder with … of FAID using a recurrent
quantized neural network (RQNN) [… that do not perform coarse non-uniform quantization, such as …

… To reduce neural network complexity and storage requirements, this paper proposes a …
, “Finite alphabet iterative decoding of LDPC codes with coarsely quantized neural networks,…

… a coarsely quantized low-density parity-check decoder that … for all decoding iterations with
a single neural network to … be extended to the decoding of non-binary LDPC codes, where the …

… QC) LDPC code show that a layered Min SumRCQ decoder with … that do not perform coarse
non-uniform quantization, such as … via recurrent quantized neural networks,” IEEE Trans. …