1: \begin{abstract}

2: To reduce the implementation complexity of a belief propagation

3: (BP) based low-density parity-check (LDPC) decoder, shuffled BP

4: decoding schedules, which serialize the decoding process by

5: dividing a complete parallel message-passing iteration into a

6: sequence of sub-iterations, have been proposed. The so-called

7: group horizontal shuffled BP algorithm partitions the check nodes

8: of the code graph into groups to perform group-by-group

9: message-passing decoding. This paper proposes a new grouping

10: technique to accelerate the message-passing rate. Performance of

11: the proposed algorithm is analyzed by a Gaussian approximation

12: approach. Both analysis and numerical experiments verify that the

13: new algorithm does yield a convergence rate faster than that of

14: existing conventional or group shuffled BP decoder with the same

15: computing complexity constraint.

16: \end{abstract}

17: