Browse by author
Lookup NU author(s): Ghanim Al-Rubaye, Professor Harris Tsimenidis, Dr Martin JohnstonORCiD
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
In this paper, we propose Irregular Non-Binary Low Density Parity Check (IR-NB-LDPC) codes with Signed Log Fast Fourier Transform (SL-FFT) decoding algorithm to overcome the harsh environment of power-line communication (PLC) channels. Their performance is compared with Irregular Binary LDPC (IR-B-LDPC) codes using the Sum Product algorithm (SPA). The sparse parity check matrix H of both codes are constructed using progressive edge growth (PEG) algorithm with a novel initialization of the apriori log likelihood ratios (LLR) of each decoder to mitigate the highly impulsive noise in PLC channels. Numerical performance results obtained via simulations show that the proposed system at hit error rate of 10(-4) achieves a coding gain of more than 21 dB compared to uncoded system and more than 6 dB compared to IR-B-LDPC codes for the same block length in bits and rates, however, this is accomplished with higher decoding complexity.
Author(s): Al-Rubaye GA, Tsimenidis CC, Johnston M
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Published
Conference Name: 23rd European Signal Processing Conference (EUSIPCO), 2015
Year of Conference: 2015
Pages: 1431-1435
Print publication date: 01/01/2015
Acceptance date: 01/01/1900
Publisher: IEEE
URL: http://dx.doi.org/10.1109/EUSIPCO.2015.7362620
DOI: 10.1109/EUSIPCO.2015.7362620
Library holdings: Search Newcastle University Library for this item
ISBN: 9780992862633