Multi Layer Perceptron Neural Networks Decoder for LDPC Codes

Karami, Alireza; Attari, Mahmoud Ahmadian; Tavakoli, Hassan

doi:10.1109/wicom.2009.5303382

Cited by 10 publications

(9 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The key problem of the construction is to find the appropriate error function. If the code word is binary, the operational relationship between the check nodes is an XOR function [12], which can be simulated by Formula (10) x…”

Section: Mlp Neural Network Decoding Structurementioning

confidence: 99%

“…In 2015, Kumar, Agrawal and Phartiyal proposed the use of artificial neural network to replace BP decoding algorithm. The binary LDPC code error was corrected through the look-up table method [12], the complexity was effectively reduced without adopting the iterative calculation. The bit flipping combination in this paper is determined by the comparison table method.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Decoding of NB-LDPC Codes based on Multi-Layer Perceptual Neural Networks

Han

Zhang

Wang

et al. 2022

Preprint

View full text Add to dashboard Cite

Low Density Parity Check (LDPC) codes have been extraordinarily important in the field of communications because its error-correction approximates to Shannon limit. Binary LDPC codes have been applied in many fields, and relatively Non-Binary (NB) LDPC codes have recently attracted considerable attention on account of their better error performance. Common NB-LDPC codes decoding algorithm involves Belief Propagation (BP) algorithm, Fast Hadamard Transform-BP (FHT-BP) algorithm, etc. The decoding performance of each algorithm has a certain upper limit. In this paper, the code word structure of NB-LDPC codes is considered, and the decoding of LDPC codes under Galois Field(4) (GF(4)) is mainly indicated. On the basis of the FHT-BP algorithm, Multi-Layer Perceptron (MLP) neural networks and bit flipping (BF) method are introduced for joint decoding. This method can improve the error performance of LDPC codes and achieve better decoding results.

show abstract

Section: Mlp Neural Network Decoding Structurementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Decoding of NB-LDPC Codes based on Multi-Layer Perceptual Neural Networks

Han

Zhang

Wang

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…According to the recent and new advances in error control coding theory, especially with using the regular [1] and irregular [2][3][4][5][6][7][8] form of Low Density Parity Check (LDPC) codes, efficient error correction schemes can be found. By using the LDPC codes and their well-known decoders such as Belief Propagation (BP) algorithm, one can achieve the whole channel capacity of the Binary Erasure Channel (BEC) [9][10][11][12] and approach to a rate, which is very close to channel capacity of Binary Symmetric Channel (BSC) [10], or Additive White Gaussian Noise Channel (AWGNC) [13].…”

Section: Introductionmentioning

confidence: 99%

Overhead minimization of online codes

Tavakoli

2018

Telecommun Syst

View full text Add to dashboard Cite

We introduce a novel method and two algorithms for designing efficient degree distributions with minimum overhead for online codes. The first algorithm is the general form for solving the problem of overhead minimization and the second one is an adaptive method based on the first method. In these two approaches, the codes are designed based on a discretized non-linear optimization problem. One direct result of the presented algorithms is decreasing the number of samples for solving the non-linear programming problem of the overhead minimization problem, which is not easy to solve without discretization. Also, we find a simple criteria in order to choose the critical sample points and show that the convergence rate improves. By considering these algorithms, one can construct an online code with minimum overhead for any given erasure channel parameter. Furthermore, the complexity of our algorithms has a linear relation with the number of samples because linear programming model is used. Simulation results are presented that show the efficiency of the proposed algorithms.

show abstract

“…Adopting neural network as main architecture of the decoder, where the neural network has relationship with the encoding rules [1][2][3][4].…”

Section: Introductionmentioning

confidence: 99%

“…Combining genetic algorithm and neural network, a genetic neural network decoder is proposed in [1], which is close to the traditional soft decision in error correction performance. However, the decoder adopts genetic algorithm as main architecture, where neural network only plays supplementary role.…”

Section: Introductionmentioning

confidence: 99%

The Design and Implementation of General Symmetrical Self-Organizing Map Decoder

Wu¹,

Ai²

2015

TOCSJ

View full text Add to dashboard Cite

Traditional error correcting decoder is designed according to specific algorithm, which is a reverse course of encoding. In this paper, a novel general neural network decoder is proposed in the form of symmetrical self-organizing map (SSOM), which can achieve decoding function to any error correcting codes. The SSOM decoder is tested by decoding low-density parity-check (LDPC) code. The performance comparison of SSOM decoder and traditional decoder is accomplished by simulation. Actual results show that the SSOM decoder can implement both learning and decoding in the same time regardless of any encoding rules. And higher possibility of the codeword emergence means greater probability of correct error correction. Compare to traditional error correcting decoder, it is easy to construct, more intelligent to different codeword sets, which has certain prospect in future communication channel coding field.

show abstract

Multi Layer Perceptron Neural Networks Decoder for LDPC Codes

Cited by 10 publications

References 5 publications

Decoding of NB-LDPC Codes based on Multi-Layer Perceptual Neural Networks

Decoding of NB-LDPC Codes based on Multi-Layer Perceptual Neural Networks

Overhead minimization of online codes

The Design and Implementation of General Symmetrical Self-Organizing Map Decoder

Contact Info

Product

Resources

About