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

International Symposium on Wireless Pervasive Computing

VLSI Architectures for Turbo Decoding Message Passing Using Min-Sum for Rate-Compatible Array LDPC Codes

Kiran Gunnam, Weihuang Wang, Gwan Choi, Mark B. Yeary

Published Jan 1, 2007• 1 citation
Publisher

Abstract

In the recent literature, turbo decoding message passing (TDMP) or layered decoding has been proposed for the decoding of low-density parity-check (LDPC) codes using a trellis-based BCJR algorithm in check node units (CNU). We present a new architecture for supporting rate compatible array LDPC codes that uses an offset-based min-sum decoding algorithm instead of the BCJR. The proposed architecture utilizes the value-reuse properties of min-sum and block-serial scheduling of computations, along with TDMP. This novel architecture has the following features: removal of memory needed to store the sum of the variable node messages and the channel values, removal of memory needed to store the variable node messages, 40%-72% savings in storage of extrinsic messages depending on rate of the codes, reduction of routers by 50%, and increase of throughput up to 2times. Implementations on our test-bed FPGA achieve decoded throughputs up to 1.36 Gbps and 2.27 Gbps for each iteration for (5,k) and (3,k) array LDPC codes, respectively. ASIC implementation achieve decoded throughputs up to 5.9 Gbps for each iteration for (5,k) array LDPC codes.

DOI
10.1109/iswpc.2007.342667

Related papers

Memory-efficient turbo decoder architectures for LDPC codes

Mohammad M. Mansour, Naresh R. Shanbhag

In this paper, we propose a turbo decoding message-passing (TDMP) algorithm to decode regular and irregular low-density parity-check (LDPC) codes. The TDMP algorithm has two main advantages over the commonly employed two-phase message-passing algorithm. First, it exhibits a faster convergence behavior (up to 50% less iterations), and improvement in coding gain (up to an order of magnitude for moderate-to-high SNR and small number of iterations). Second, the corresponding decoder architecture has a significantly reduced memory requirement that amounts to a savings of (75 + 25n//spl Sigma/ node-degrees)% > 75% for code-length n. A decoder architecture featuring the TDNW algorithm is also presented. Furthermore, we propose a new structure on the parity-check matrix of an LDPC code based on permutation matrices aimed at reducing interconnect complexity and improving decoding throughput. In addition, we construct a wide range of LDPC codes based on Ramanujan graphs which possess this structure.

SiPS ... design and implementation - IEEE Workshop on Signal Processing Systems•Aug 27, 2003•35 citations

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

Design and implementation of decoders for error correction in high-speed communication systems.

Joan Marc Català Pérez

This thesis is focused on the design and implementation of binary low-density parity-check (LDPC) code decoders for high-speed modern communication systems. The basic of LDPC codes and the performance and bottlenecks, in terms of complexity and hardware efficiency, of the main soft-decision and hard-decision decoding algorithms (such as Min-Sum, Optimized 2-bit Min-Sum and Reliability-based iterative Majority-Logic) are analyzed. The complexity and performance of those algorithms are improved to allow efficient hardware architectures.\n\nA new decoding algorithm called One-Minimum Min-Sum is proposed. It reduces considerably the complexity of the check node update equations of the Min-Sum algorithm. The second minimum is estimated from the first minimum value by a means of a linear approximation that allows a dynamic adjustment. The Optimized 2-bit Min-Sum algorithm is modified to initialize it with the complete LLR values and to introduce the extrinsic information in the messages sent from the variable nodes. Its variable node equation is reformulated to reduce its complexity. Both algorithms were tested for the (2048,1723) RS-based LDPC code and (16129,15372) LDPC code using an FPGA-based hardware emulator. They exhibit BER performance very close to Min-Sum algorithm and do not introduce early error-floor. \n\nIn order to show the hardware advantages of the proposed algorithms, hardware decoders were implemented in a 90 nm CMOS process and FPGA devices based on two types of architectures: full-parallel and partial-parallel one with horizontal layered schedule. The results show that the decoders are more area-time efficient than other published decoders and that the low-complexity of the Modified Optimized 2-bit Min-Sum allows the implementation of 10 Gbps decoders in current FPGA devices. \n\nFinally, a new hard-decision decoding algorithm, the Historical-Extrinsic Reliability-Based Iterative Decoder, is presented. This algorithm introduces the new idea of considering hard-decision votes as soft-decision to compute the extrinsic information of previous iterations. It is suitable for high-rate codes and improves the BER performance of the previous RBI-MLGD algorithms, with similar complexity.

•Jul 7, 2017•0 citations

Decoding of Quasi-cyclic LDPC Codes Using an On-the-Fly Computation

Kiran Gunnam, Gwan Choi, Weihuang Wang +2 more

The implementation complexity of the decoder for low-density parity-check codes (LDPC) is dictated by memory and interconnect requirements. We propose new LDPC decoder architectures that reduce the need of message passing memory by 80% (for standard message passing)-55%(for layered decoding) and the router requirements by more than 50%. These novel architectures are based on scheduling of computation that results in "on the fly computation" of variable node and check node reliability messages. These architectures are targeted for quasi- cyclic LDPC codes such as array LDPC codes (regular QC-LDPC) and block LDPC codes (irregular QC-LDPC). FPGA and ASIC implementation results show substantial gains when compared to the existing work in the literature.

Asilomar Conference on Signals, Systems and Computers (ACSSC)•Jan 1, 2006•26 citations

Efficient Message Passing Architecture for High Throughput LDPC Decoder

Zhiqiang Cui, Zhongfeng Wang

In this paper, we propose an efficient message passing architecture for permutation matrices based LDPC code decoders. Min-sum algorithm is reformulated to facilitate significant reduction of routing complexity and memory usage. For a (2048, 1723) (6, 32) LDPC code with 4-bit quantization, 54% outgoing wires per variable node unit and 90% outgoing wires per check node unit can be saved. To further reduce hardware complexity, an optimized nonuniform quantization scheme using only 3 bits to represent each message has been investigated. The simulation result shows that it has only 0.25dB performance loss from the floating-point SPA

IEEE International Symposium on Circuits and Systems proceedings•May 1, 2007•14 citations

Reduced-complexity extended Min-sum check node processing for non-binary LDPC decoding

Xinmiao Zhang, Fang Cai

Non-binary low-density parity-check (NB-LDPC) codes can achieve higher coding gain than binary LDPC codes when the code length is moderate at the cost of higher complexity. One major step of NB-LDPC decoding is check node processing. Previously, iterative forward-backward approaches are employed to implement this step. However, the storage of the intermediate results of the forward and backward computations requires large memory. In this paper, a novel check node processing scheme and corresponding VLSI architectures are proposed for the extended Min-sum NB-LDPC decoding. The proposed scheme only stores a limited number of sorted variable-to-check messages. Then the check-to-variable messages for different variable node are generated independently. For a (837, 726) NB-LDPC code over GF(25), the proposed architecture only requires 64% of the area of the previous design with the same throughput and error-correcting performance.

IEEE International Midwest Symposium on Circuits and Systems (MWSCAS)•Aug 1, 2010•5 citations
In the recent literature, turbo decoding message passing (TDMP) or layered decoding has been proposed for the decoding of low-density parity-check (LDPC) codes using a trellis-based BCJR algorithm in check node units (CNU). We present a new architecture for supporting rate compatible array LDPC codes that uses an offset-based min-sum decoding algorithm instead of the BCJR. The proposed architecture utilizes the value-reuse properties of min-sum and block-serial scheduling of computations, along with TDMP. This novel architecture has the following features: removal of memory needed to store the sum of the variable node messages and the channel values, removal of memory needed to store the variable node messages, 40%-72% savings in storage of extrinsic messages depending on rate of the codes, reduction of routers by 50%, and increase of throughput up to 2times. Implementations on our test-bed FPGA achieve decoded throughputs up to 1.36 Gbps and 2.27 Gbps for each iteration for (5,k) and (3,k) array LDPC codes, respectively. ASIC implementation achieve decoded throughputs up to 5.9 Gbps for each iteration for (5,k) array LDPC codes.
In this paper, we propose a turbo decoding message-passing (TDMP) algorithm to decode regular and irregular low-density parity-check (LDPC) codes. The TDMP algorithm has two main advantages over the commonly employed two-phase message-passing algorithm. First, it exhibits a faster convergence behavior (up to 50% less iterations), and improvement in coding gain (up to an order of magnitude for moderate-to-high SNR and small number of iterations). Second, the corresponding decoder architecture has a significantly reduced memory requirement that amounts to a savings of (75 + 25n//spl Sigma/ node-degrees)% > 75% for code-length n. A decoder architecture featuring the TDNW algorithm is also presented. Furthermore, we propose a new structure on the parity-check matrix of an LDPC code based on permutation matrices aimed at reducing interconnect complexity and improving decoding throughput. In addition, we construct a wide range of LDPC codes based on Ramanujan graphs which possess this structure.
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 thesis is focused on the design and implementation of binary low-density parity-check (LDPC) code decoders for high-speed modern communication systems. The basic of LDPC codes and the performance and bottlenecks, in terms of complexity and hardware efficiency, of the main soft-decision and hard-decision decoding algorithms (such as Min-Sum, Optimized 2-bit Min-Sum and Reliability-based iterative Majority-Logic) are analyzed. The complexity and performance of those algorithms are improved to allow efficient hardware architectures.\n\nA new decoding algorithm called One-Minimum Min-Sum is proposed. It reduces considerably the complexity of the check node update equations of the Min-Sum algorithm. The second minimum is estimated from the first minimum value by a means of a linear approximation that allows a dynamic adjustment. The Optimized 2-bit Min-Sum algorithm is modified to initialize it with the complete LLR values and to introduce the extrinsic information in the messages sent from the variable nodes. Its variable node equation is reformulated to reduce its complexity. Both algorithms were tested for the (2048,1723) RS-based LDPC code and (16129,15372) LDPC code using an FPGA-based hardware emulator. They exhibit BER performance very close to Min-Sum algorithm and do not introduce early error-floor. \n\nIn order to show the hardware advantages of the proposed algorithms, hardware decoders were implemented in a 90 nm CMOS process and FPGA devices based on two types of architectures: full-parallel and partial-parallel one with horizontal layered schedule. The results show that the decoders are more area-time efficient than other published decoders and that the low-complexity of the Modified Optimized 2-bit Min-Sum allows the implementation of 10 Gbps decoders in current FPGA devices. \n\nFinally, a new hard-decision decoding algorithm, the Historical-Extrinsic Reliability-Based Iterative Decoder, is presented. This algorithm introduces the new idea of considering hard-decision votes as soft-decision to compute the extrinsic information of previous iterations. It is suitable for high-rate codes and improves the BER performance of the previous RBI-MLGD algorithms, with similar complexity.
The implementation complexity of the decoder for low-density parity-check codes (LDPC) is dictated by memory and interconnect requirements. We propose new LDPC decoder architectures that reduce the need of message passing memory by 80% (for standard message passing)-55%(for layered decoding) and the router requirements by more than 50%. These novel architectures are based on scheduling of computation that results in "on the fly computation" of variable node and check node reliability messages. These architectures are targeted for quasi- cyclic LDPC codes such as array LDPC codes (regular QC-LDPC) and block LDPC codes (irregular QC-LDPC). FPGA and ASIC implementation results show substantial gains when compared to the existing work in the literature.
In this paper, we propose an efficient message passing architecture for permutation matrices based LDPC code decoders. Min-sum algorithm is reformulated to facilitate significant reduction of routing complexity and memory usage. For a (2048, 1723) (6, 32) LDPC code with 4-bit quantization, 54% outgoing wires per variable node unit and 90% outgoing wires per check node unit can be saved. To further reduce hardware complexity, an optimized nonuniform quantization scheme using only 3 bits to represent each message has been investigated. The simulation result shows that it has only 0.25dB performance loss from the floating-point SPA
Non-binary low-density parity-check (NB-LDPC) codes can achieve higher coding gain than binary LDPC codes when the code length is moderate at the cost of higher complexity. One major step of NB-LDPC decoding is check node processing. Previously, iterative forward-backward approaches are employed to implement this step. However, the storage of the intermediate results of the forward and backward computations requires large memory. In this paper, a novel check node processing scheme and corresponding VLSI architectures are proposed for the extended Min-sum NB-LDPC decoding. The proposed scheme only stores a limited number of sorted variable-to-check messages. Then the check-to-variable messages for different variable node are generated independently. For a (837, 726) NB-LDPC code over GF(25), the proposed architecture only requires 64% of the area of the previous design with the same throughput and error-correcting performance.