1. A Genetic Algorithm Decoder for Low-Density Parity-Check Codes Over an Impulse Noise Channel
- Author
-
Ramiro Avalos Ribas, Ivan Exequiel Gelosi, Monica Cristina Liberatori, Jorge Castineira Moreira, L. Coppolillo, and Alejandro José Uriz
- Subjects
Computer science ,020208 electrical & electronic engineering ,Word error rate ,02 engineering and technology ,Communications system ,Impulse noise ,Noise ,Genetic algorithm ,0202 electrical engineering, electronic engineering, information engineering ,Low-density parity-check code ,Algorithm ,Decoding methods ,Computer Science::Information Theory ,Communication channel - Abstract
In this paper, a decoder for Low-Density ParityCheck codes is proposed. The system employs a genetic algorithm aided by syndrome weight selection and distance logic. Genetic algorithms were chosen due to their capability to perform heuristic searches using evolution-based convergence while also exploring wide spaces. The paper has considered the performance of the decoder over an impulse noise channel, typical to power line communications, and a type of disturbance commonly found in most transmission mediums and devices. The resulting system is shown to perform as well as the standard sum-product decoder, with the additional advantage of not requiring knowledge of the channel information (noise type or power). A parallel decoding topology is proposed, where error rate performance can be improved by the addition of concurrent decoders. Furthermore, this scheme allows for genetic algorithm complexity to be reduced, as the addition of parallel blocks can compensate the individual loss in error rate performance.
- Published
- 2019
- Full Text
- View/download PDF