IEEE International Symposium on Information Theory (ISIT)

Performance Bounds for Quantized Spatially Coupled LDPC Decoders Based on Absorbing Sets

Homayoon Hatami, David G. M. Mitchell, Daniel J. Costello, Thomas E. Fuja

Published Jun 1, 2018• 7 citations

Abstract

Absorbing sets are known to be the primary factor in the error-floor performance of low-density parity-check (LDPC) codes with message passing decoders over the additive white Gaussian noise (AWGN) channel. Besides showing excellent waterfall performance, spatially coupled LDPC (SC-LDPC) codes that are constructed by an edge spreading technique are known to have fewer cycles and absorbing sets than their block code counterparts, and therefore to exhibit better error-floor performance. Based on our previously obtained results for quantized LDPC block decoders, we derive lower bounds on the performance of quantized SC-LDPC decoders, including both a flooding schedule decoder and a sliding window decoder. Numerical simulation results confirm the accuracy of the obtained bounds and show that, for quantized decoders, properly designed SC-LDPC codes have better error-floor performance than their underlying LDPC block codes.

DOI: 10.1109/isit.2018.8437710

Related papers

Edge spreading design of high rate array-based SC-LDPC codes

David G. M. Mitchell, Eirik Rosnes

Absorbing sets (ASs) are combinatorially defined objects existing in the Tanner graph of a low-density parity-check (LDPC) code that have been shown to cause failures in the iterative message-passing decoder when transmission occurs over the additive white Gaussian noise channel. In this paper, we propose an edge spreading approach to construct high rate array-based spatially-coupled LDPC codes by jointly optimizing the AS spectrum and the minimum distance. By considering general edge spreadings and by considering a larger memory, we show that strictly better codes can be constructed, both in terms of achievable minimum distance for small-to-moderate block lengths and in terms of the number of small ASs.

IEEE International Symposium on Information Theory (ISIT)Jun 1, 201727 citations

Lower bounds for quantized LDPC min-sum decoders based on absorbing sets

Homayoon Hatami, David G. M. Mitchell, Daniel J. Costello +1 more

Previously published lower bounds on the error-floor performance of low-density parity-check (LDPC) codes with quantized sum-product algorithm (SPA) decoders need modification to apply to min-sum algorithm (MSA) decoders. In this paper we show that, due to the sub-optimality of MSA decoders, certain assumptions that apply to SPA decoders are not valid for MSA decoders. Based on appropriately modified assumptions, an improved technique is then proposed to evaluate lower bounds on the error floor performance of LDPC codes for an additive white Gaussian noise channel and a quantized MSA decoder. This new technique not only results in error floor bounds for MSA decoders, but also for SPA decoders. The obtained bound is based on an absorbing set, and a good error floor approximation can be obtained when the multiplicity of the absorbing set is known.

Allerton Conference on Communication, Control, and ComputingOct 1, 20174 citations

Absorbing set characterization of array-based spatially coupled LDPC codes

David G. M. Mitchell, Lara Dolecek, Daniel J. Costello

Absorbing sets are combinatorially defined objects existing in the Tanner graph of a low-density parity-check (LDPC) code that have been shown to cause failures in the iterative message-passing decoder when transmission occurs over the additive white Gaussian noise channel. In this paper, we study the absorbing set properties of a class of high-rate array-based spatially coupled LDPC (SC-LDPC) codes that are constructed by coupling together L array-based LDPC block codes. We prove that the smallest absorbing sets existing in the Tanner graph of the SC-LDPC code have the same size as those in the corresponding uncoupled LDPC codes, and the number of such sets grow linearly with L. We show that spatial coupling greatly reduces the average number (per symbol) of minimal sets compared to the uncoupled codes, and we explain that this reduction is due to many absorbing sets and small cycles being `broken' by the coupling process. The large reduction in the number of minimal absorbing sets suggests that array-based SC-LDPC codes will have significantly improved decoding performance in the high signal-to-noise ratio regime compared to the corresponding uncoupled LDPC codes.

IEEE International Symposium on Information Theory (ISIT)Jun 1, 201444 citations

An Absorbing Set Analysis of Quantized LDPC Decoders

Homayoon Hatami

The performance of low-density parity-check (LDPC) codes at high signal-to- noise ratios (SNRs) is known to be limited by the presence of certain sub-graphs that exist in the Tanner graph representation of the code, for example trapping sets and absorbing sets. This dissertation derives a lower bound on the frame error rate (FER) of any LDPC code containing a given problematic sub-graph, assuming a particular iterative message passing (MP) decoder and decoder quantization. A crucial aspect of the lower bound is that it is code-independent, in the sense that it can be derived based only on a problematic sub-graph and then applied to any code containing it. Due to the complexity of evaluating the exact bound, assumptions are proposed to approximate it, from which we can estimate decoder performance. For LDPC codes, the attenuated min-sum algorithm (AMSA) and the offset min- sum algorithm (OMSA) are known to outperform the conventional min-sum algorithm (MSA) at low SNRs, i.e., in the waterfall region of the FER curve. This dis- sertation demonstrates that, for quantized LDPC decoders, the AMSA and OMSA underperform the MSA at high SNRs, i.e., in the error floor region of the FER curve, and that the MSA, AMSA, and OMSA all suffer from a relatively high error floor. Therefore, we introduce a novel modification of the MSA for quantized decoding of LDPC codes, called the threshold attenuated min-sum algorithm (TAMSA) (or the threshold offset min-sum algorithm (TOMSA)) with the aim of outperforming the MSA, AMSA, and OMSA at both low and high SNRs. The new decoding algorithm is designed based on the common assumption that trapping sets are the major cause of the error floor for quantized LDPC decoders, and it is shown to reduce the error probability associated with these problematic graphic objects. The lower bound analysis tool noted above is then used to verify the effectiveness of the new TAMSA (TOMSA) and also to assist in finding its optimum design parameters. We also show that the implementation complexity of the TAMSA (TOMSA) is only slightly higher than that of the AMSA or OMSA. Finally, we extend the lower bound to sliding window (SW) decoding of spatially coupled LDPC (SC-LDPC) codes. Additionally, we show that the performance of the TAMSA (TOMSA) is better than that of the MSA, AMSA, and OMSA for SW.

Nov 16, 20220 citations

Asymptotic and finite-length performance of irregular spatially-coupled codes

Reza A. Ashrafi, Abdullah Sarıduman, Alí Emre Pusane

The newest family of low-density parity-check (LDPC) codes, spatially-coupled (SC) codes, is shown to have several desirable characteristics including low implementation complexity and close-to-optimal performance over a range of channels. Furthermore, because of their ribbon-shaped parity-check matrices, window decoding can be used to decode these codes, which leads to low-delay implementations. Researchers have focused on asymptotically regular SC code ensembles and have examined several aspects of the code construction processes. In this paper, we concentrate on irregular SC code ensembles. We evaluate their decoding thresholds over the binary erasure channel and show that their performance is better than their regular SC counterparts. It is also shown that the gap between asymptotic coding thresholds of irregular SC ensembles and the fundamental Shannon limit gets negligibly small. For the sake of a better comparison, we have also evaluated the finite-length error performance of selected regular and irregular SC codes over the additive white Gaussian channel and it is also observed that finite-length error performance of these irregular SC codes outperforms regular SC codes. To further improve the error performance of these codes and to lower the possible error floors, progressive edge growth algorithm has also been considered in the finite-length performance analysis.

IEEE International Black Sea Conference on Communications and NetworkingMay 1, 20141 citation

Performance bounds for quantized LDPC decoders based on absorbing sets

Homayoon Hatami, David G. M. Mitchell, Daniel J. Costello +1 more

A code-independent performance bound for a given absorbing set is derived for quantized low-density parity-check (LDPC) decoders. The analysis demonstrates that each absorbing set in the Tanner graph imposes a specific lower bound on the frame error rate (FER) of any code containing that absorbing set under a given quantization scheme. This approach is applicable to any message-passing (MP) decoding algorithm and any uniform or non-uniform quantization scheme for LDPC codes. Simulation results using the sum-product algorithm (SPA) provide FERs that are consistent with the obtained bounds. In addition, the bounds demonstrate that the conventional quantized SPA is not capable of achieving very low FERs if the LDPC codes contain certain absorbing sets.

IEEE International Symposium on Information Theory (ISIT)Jul 1, 20165 citations