Back to Search
Start Over
Machine learning for decoding linear block codes: case of multi-class logistic regression model
- Publication Year :
- 2021
- Publisher :
- Zenodo, 2021.
-
Abstract
- Facing the challenge of enormous data sets variety, several machine learning-based algorithms for prediction (e.g, Support vector machine, multi layer perceptron and logistic regression) have been highly proposed and used over the last years in many fields. Error correcting codes (ECCs) are extensively used in practice to protect data against damaged data storage systems and against random errors due to noise effects. In this paper, we will use machine learning methods, especially multi-class logistic regression combined with the famous syndrome decoding algorithm. The main idea behind our decoding method which we call logistic regression decoder (LRDec) is to use the efficient multi-class logistic regression models to find errors from syndromes in linear codes such as bose, ray-chaudhuri and hocquenghem (BCH), and the quadratic residue (QR). Obtained results of the proposed decoder have a significant benefit in terms of bit error rate (BER) for random binary codes. The comparison of our decoder with many competitors proves its power. The proposed decoder has reached a success percentage of 100% for correctable errors in the studied codes.
- Subjects :
- Control and Optimization
Computer Networks and Communications
Computer science
Logistic regression
Data_CODINGANDINFORMATIONTHEORY
Machine learning
computer.software_genre
BCH codes
Error correction codes
Syndrome decoding
Electrical and Electronic Engineering
business.industry
SoftMax
Linear code
Support vector machine
Hardware and Architecture
Multilayer perceptron
Signal Processing
Softmax function
Binary code
Artificial intelligence
business
computer
BCH code
Decoding methods
Information Systems
Subjects
Details
- Language :
- English
- Database :
- OpenAIRE
- Accession number :
- edsair.doi.dedup.....b616d0666f22a8d0462d9fb396c7395a