LDPC Decoders with Informed Dynamic Scheduling

Abstract

Abstract-Low-Density Parity-Check (LDPC) codes are usually decoded by running an iterative belief-propagation (BP), or message-passing, algorithm over the factor graph of the code. The traditional message-passing scheduling, called flooding, consists of updating all the variable nodes in the graph, using the same pre-update information, followed by updating all the check nodes of the graph, again, using the same pre-update information. Recently, several studies show that sequential scheduling, in which messages are generated using the latest available information, significantly improves the convergence speed in terms of number of iterations. Sequential scheduling introduces the problem of finding the best sequence of message updates. We propose Informed Dynamic Scheduling (IDS) strategies that select the message-passing schedule according to the observed rate of change of the messages. In general, IDS strategies require computation to select the message to update but converge in fewer message updates because they focus on the part of the graph that has not converged. Moreover, IDS yields a lower errorrate performance than either flooding or sequential scheduling because IDS strategies overcome traditional trapping-set errors. This paper presents IDS strategies that address several issues including performance for short-blocklength codes, complexity, and implementability.

Augmented decoders for LDPC codes

“…This code has similar construction to the WiMAX code. Here, augmentation is compared to the performance of the approximate node-wise scheduling (ANS) serial decoder presented in [12]. It can be seen in Fig.…”

“…If the BP messages are limited to a small number of bits, then modifying the quantization scheme can improve the performance [11]. The scheduling of message passing in a BP decoder can also be altered, giving rise to serial decoders [12][13][14]. It is also possible to identify particular trapping sets and apply an appropriate perturbation if a known trapping set is present [15].…”

Effective lazy schedule for the layered belief propagation algorithm

“…The LS [9], informed dynamic schedule (IDS) [10], and its transformations [11][12][13][14][15][16] have shown that most of the variable nodes would achieve high confidence after a few iterations. The LS adopts the strategy that the related variable nodes will be set to sleep mode once they are reliable enough, and those variable nodes will awake when the conditions are broken.…”

“…Recently, a residual belief propagation (RBP) [10] was raised, and its transformations (e.g. approximate node-wise RBP (ANS-RBP) algorithms [11], variable node-wise RBP (V-RBP) [12], variable-to-check RBP (VC-RPB) [13] and informed VC-RBP (IVC-RBP) [14]) were presented. Huang and Zhang [15] introduced two kinds of relative-RBP algorithms (CN-RRDS and VN-RRDS).…”

“…This paper presents such an effort and associated results. Inspired by the RBP and its developments [10][11][12][13][14][15][16], a LS for the LBP (LS-LBP) is proposed. In the LBP algorithm, the messages are updated as a determinate sequence of check nodes, and the latter update will use the previous updated messages immediately.…”