Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

European Signal Processing Conference (EUSIPCO)

Fully Programmable Layered Ldpc Decoder Architecture

Christiane Beuschel, Pfleiderer, Hans-Jörg

Published Aug 24, 2009• 0 citations
Publisher

Abstract

In this article we present a fully programmable layered LDPC decoder architecture together with an optimum mapping and scheduling algorithm. In contrast to other designs proposed in the literature, we use one-phase message passing. This allows for the first time the design of a fully programmable layered decoder. The proposed mapping and scheduling algorithm exploits the full parallelism of the architecture at any time for any code, which means that the mapping algorithm achieves collision-free memory access and 100 % utilization of the architecture. Compared to existing programmable designs without layered decoding we double the data throughput. The parallelism of the architecture is unconstrained and fully scalable so that hardware cost and data throughput can be exchanged with fine granularity. 1.

DOI
10.5281/zenodo.41409

Related papers

Fully programmable decoder architecture for structured and unstructured LDPC codes

Christiane Beuschel, Hans-Jörg Pfleiderer

In this article we present a fully programmable and scalable partly-parallel LDPC decoder architecture together with an optimum mapping and scheduling algorithm. The proposed algorithm exploits the full parallelism of the architecture at any time for any code, which means that the mapping algorithm achieves 100% utilization of the architecture. The proposed design is fully programmable and can be reconfigured for a different LDPC code by changing the initialization of the control memory. Thus the architecture can be used for a multi-standard decoder which supports decoding of any structured or unstructured LDPC code. Furthermore, the parallelism of the architecture is unconstrained and fully scalable which allows to exchange hardware cost and throughput with fine granularity. In contrast to previously proposed programmable designs our approach uses parallel variable and check node processing and thus doubles the data throughput.

International Conference on Wireless Communication, Vehicular Technology, Information Theory and Aerospace & Electronic Systems Technology•May 1, 2009•5 citations

Fully programmable LDPC decoder hardware architectures

Christiane Beuschel

In recent years, the amount of digital data which is stored and transmitted for private and public usage has increased considerably. To allow a save transmission and storage of data despite of error-prone transmission media, error correcting codes are used. A large variety of codes has been developed, and in the past decade low-density parity-check (LDPC) codes which have an excellent error correction performance became more and more popular. Today, low-density parity-check codes have been adopted for several standards, and efficient decoder hardware architectures are known for the chosen structured codes. However, the existing decoder designs lack flexibility as only few structured codes can be decoded with one decoder chip. In consequence, different codes require a redesign of the decoder, and few solutions exist for decoding of codes which are not quasi-cyclic or which are unstructured. In this thesis, three different approaches are presented for the implementation of fully programmable LDPC decoders which can decode arbitrary LDPC codes. As a design study, the first programmable decoder which uses a heuristic mapping algorithm is realized on a field-programmable gate array (FPGA), and error correction curves are measured to verify the correct functionality. The main contribution of this thesis lies in the development of the second and the third architecture and an appropriate mapping algorithm. The proposed fully programmable decoder architectures use one-phase message passing and layered decoding and can decode arbitrary LDPC codes using an optimum mapping and scheduling algorithm. The presented programmable architectures are in fact generalized decoder architectures from which the known decoder architectures for structured LDPC codes can be derived.

•Jan 1, 2010•1 citation

High throughput layered decoding of LDPC codes

Mohamed Ismail, Justin P. Coon, Imran Ahmed +2 more

Layered decoding of Low Density Parity Check (LDPC) codes has been shown to give faster convergence and use less storage than the conventional Two Phase Message Passing (TPMP) algorithm in decoding such codes. To achieve higher throughput, multiple rows of the parity check matrix may be decoded in parallel, as in the Parallel-Turbo Decoding Message Passing (P-TDMP) algorithm. However, contention for globally shared memory permits only a subset of rows to be decoded in parallel thus limiting overall throughput. This paper uses a near optimal scheduling algorithm, based on graph colouring, to overcome memory contention, allowing processing of multiple rows without code modification, resulting in significant throughput gains over the standard P-TDMP algorithm. For the LDPC codes studied, throughput gains compared to the P-TDMP algorithm, of between ten and fifteen are shown to be possible.

IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)•Sep 1, 2009•0 citations

Ultra High Throughput Unrolled Layered Architecture for QC-LDPC Decoders

Oana Boncalo, Alexandru Amaricăi

This paper proposes a layered decoder architecture for array QC-LDPC codes which targets tens of Gbps data rates. It relies on layer unrolling with pipeline stages in between layers, allowing simultaneous decoding of multiple layers. The most important features of the proposed decoder are: (i) fully parallel processing units within each layer (ii) hardwired layer interconnect that allows the removal of high cost variable shift units, (iii) A posteriori log-likelihood ratio (AP-LLR) message memory type of storage is replaced by in-between layers pipeline registers, as the messages are being forwarded from one layer to the next. Data dependencies is this masively parallel structure is resolved by decoding multi-codewords at time. Hence, the proposed architecture allows optimum throughput/cost ratio. FPGA based implementation results indicate that for a 1296 bits LDPC code with 3 layers, throughput of up to 62 Gbps for an average of 4 iterations is by obtained using the proposed architecture.

IEEE Computer Society Annual Symposium on VLSI•Jul 1, 2017•16 citations

Decoder architecture for array-code-based LDPC codes

S. Ölçer

We describe a decoder architecture intended for decoding array-code-based low-density parity-check (LDPC) codes using the sum-product algorithm (SPA). The advantages of the proposed architecture, as compared to the fully parallel implementation of the SPA, are: reduced memory size, avoidance of complex signal interconnect patterns, and ease of programmability to accommodate various code parameters. These advantages are derived from exploiting the well-defined structure of the parity-check matrix of array-code based LDPC codes. Sum-product decoding with modified message-passing schedules are proposed to simplify the decoder implementation further.

Globecom•Mar 22, 2004•47 citations

Scheduling Algorithm for Partially Parallel Architecture of LDPC Decoder by Matrix Permutation

In‐Cheol Park, Se-Hyeon Kang

The fully parallel LDPC decoding architecture can achieve high decoding throughput, but it suffers from large hardware complexity caused by a large set of processing units (PUs) and complex interconnections. A practical solution of area-efficient decoders is to use the partially parallel architecture in which a PU is shared for several rows or columns. It is important in the partially parallel architecture to determine the rows or columns to be processed in a PU and their processing order. The dependencies between rows and columns should be considered to minimize the overall processing time by overlapping the decoding operations. This paper proposes an efficient scheduling algorithm that can be applied to general LDPC codes, which is based on the concept of matrix permutation. Experimental results show that the proposed scheduling achieves a higher decoding rate, leading to a reduction of 25% processing time on average. A 1024-bit rate-1/2 LDPC decoder, employing the proposed scheduling algorithm, provides almost 1 Gbps decoding throughput and occupies one-fifth the area compared to the fully parallel decoder.

IEEE International Symposium on Circuits and Systems proceedings•Jul 27, 2005•27 citations
In this article we present a fully programmable layered LDPC decoder architecture together with an optimum mapping and scheduling algorithm. In contrast to other designs proposed in the literature, we use one-phase message passing. This allows for the first time the design of a fully programmable layered decoder. The proposed mapping and scheduling algorithm exploits the full parallelism of the architecture at any time for any code, which means that the mapping algorithm achieves collision-free memory access and 100 % utilization of the architecture. Compared to existing programmable designs without layered decoding we double the data throughput. The parallelism of the architecture is unconstrained and fully scalable so that hardware cost and data throughput can be exchanged with fine granularity. 1.
In this article we present a fully programmable and scalable partly-parallel LDPC decoder architecture together with an optimum mapping and scheduling algorithm. The proposed algorithm exploits the full parallelism of the architecture at any time for any code, which means that the mapping algorithm achieves 100% utilization of the architecture. The proposed design is fully programmable and can be reconfigured for a different LDPC code by changing the initialization of the control memory. Thus the architecture can be used for a multi-standard decoder which supports decoding of any structured or unstructured LDPC code. Furthermore, the parallelism of the architecture is unconstrained and fully scalable which allows to exchange hardware cost and throughput with fine granularity. In contrast to previously proposed programmable designs our approach uses parallel variable and check node processing and thus doubles the data throughput.
In recent years, the amount of digital data which is stored and transmitted for private and public usage has increased considerably. To allow a save transmission and storage of data despite of error-prone transmission media, error correcting codes are used. A large variety of codes has been developed, and in the past decade low-density parity-check (LDPC) codes which have an excellent error correction performance became more and more popular. Today, low-density parity-check codes have been adopted for several standards, and efficient decoder hardware architectures are known for the chosen structured codes. However, the existing decoder designs lack flexibility as only few structured codes can be decoded with one decoder chip. In consequence, different codes require a redesign of the decoder, and few solutions exist for decoding of codes which are not quasi-cyclic or which are unstructured. In this thesis, three different approaches are presented for the implementation of fully programmable LDPC decoders which can decode arbitrary LDPC codes. As a design study, the first programmable decoder which uses a heuristic mapping algorithm is realized on a field-programmable gate array (FPGA), and error correction curves are measured to verify the correct functionality. The main contribution of this thesis lies in the development of the second and the third architecture and an appropriate mapping algorithm. The proposed fully programmable decoder architectures use one-phase message passing and layered decoding and can decode arbitrary LDPC codes using an optimum mapping and scheduling algorithm. The presented programmable architectures are in fact generalized decoder architectures from which the known decoder architectures for structured LDPC codes can be derived.
Layered decoding of Low Density Parity Check (LDPC) codes has been shown to give faster convergence and use less storage than the conventional Two Phase Message Passing (TPMP) algorithm in decoding such codes. To achieve higher throughput, multiple rows of the parity check matrix may be decoded in parallel, as in the Parallel-Turbo Decoding Message Passing (P-TDMP) algorithm. However, contention for globally shared memory permits only a subset of rows to be decoded in parallel thus limiting overall throughput. This paper uses a near optimal scheduling algorithm, based on graph colouring, to overcome memory contention, allowing processing of multiple rows without code modification, resulting in significant throughput gains over the standard P-TDMP algorithm. For the LDPC codes studied, throughput gains compared to the P-TDMP algorithm, of between ten and fifteen are shown to be possible.
This paper proposes a layered decoder architecture for array QC-LDPC codes which targets tens of Gbps data rates. It relies on layer unrolling with pipeline stages in between layers, allowing simultaneous decoding of multiple layers. The most important features of the proposed decoder are: (i) fully parallel processing units within each layer (ii) hardwired layer interconnect that allows the removal of high cost variable shift units, (iii) A posteriori log-likelihood ratio (AP-LLR) message memory type of storage is replaced by in-between layers pipeline registers, as the messages are being forwarded from one layer to the next. Data dependencies is this masively parallel structure is resolved by decoding multi-codewords at time. Hence, the proposed architecture allows optimum throughput/cost ratio. FPGA based implementation results indicate that for a 1296 bits LDPC code with 3 layers, throughput of up to 62 Gbps for an average of 4 iterations is by obtained using the proposed architecture.
We describe a decoder architecture intended for decoding array-code-based low-density parity-check (LDPC) codes using the sum-product algorithm (SPA). The advantages of the proposed architecture, as compared to the fully parallel implementation of the SPA, are: reduced memory size, avoidance of complex signal interconnect patterns, and ease of programmability to accommodate various code parameters. These advantages are derived from exploiting the well-defined structure of the parity-check matrix of array-code based LDPC codes. Sum-product decoding with modified message-passing schedules are proposed to simplify the decoder implementation further.
The fully parallel LDPC decoding architecture can achieve high decoding throughput, but it suffers from large hardware complexity caused by a large set of processing units (PUs) and complex interconnections. A practical solution of area-efficient decoders is to use the partially parallel architecture in which a PU is shared for several rows or columns. It is important in the partially parallel architecture to determine the rows or columns to be processed in a PU and their processing order. The dependencies between rows and columns should be considered to minimize the overall processing time by overlapping the decoding operations. This paper proposes an efficient scheduling algorithm that can be applied to general LDPC codes, which is based on the concept of matrix permutation. Experimental results show that the proposed scheduling achieves a higher decoding rate, leading to a reduction of 25% processing time on average. A 1024-bit rate-1/2 LDPC decoder, employing the proposed scheduling algorithm, provides almost 1 Gbps decoding throughput and occupies one-fifth the area compared to the fully parallel decoder.