A. Kharin, Igor Volkov, Evgeny Mirokhin, Konstantin Terekhov, Aleksei Dryakhlov, K. Zavertkin, A. Ovinnikov, E. Likhobabin, V. Vityazev
{"title":"Parallel LDPC decoder implementation on a mobile GPU","authors":"A. Kharin, Igor Volkov, Evgeny Mirokhin, Konstantin Terekhov, Aleksei Dryakhlov, K. Zavertkin, A. Ovinnikov, E. Likhobabin, V. Vityazev","doi":"10.1109/TELFOR.2017.8249333","DOIUrl":null,"url":null,"abstract":"In this paper we explore acceleration of the belief propagation (BP) and BP-based algorithms for decoding LDPC codes, using parallel computation on graphics processing units (GPUs) for mobile hardware platform. Different modifications of BP decoding algorithm are considered in search for best error correction performance in limited decoding time. Possible combinations of decoding algorithms, number precision and compute resources were found. Error correction performance for each found combination was measured. Best combination of decoding algorithm, number precision and compute resource was found for data transmission system with LDPC coding for DVB-T2/S2 (16200, 7200) code.","PeriodicalId":422501,"journal":{"name":"2017 25th Telecommunication Forum (TELFOR)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 25th Telecommunication Forum (TELFOR)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/TELFOR.2017.8249333","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
In this paper we explore acceleration of the belief propagation (BP) and BP-based algorithms for decoding LDPC codes, using parallel computation on graphics processing units (GPUs) for mobile hardware platform. Different modifications of BP decoding algorithm are considered in search for best error correction performance in limited decoding time. Possible combinations of decoding algorithms, number precision and compute resources were found. Error correction performance for each found combination was measured. Best combination of decoding algorithm, number precision and compute resource was found for data transmission system with LDPC coding for DVB-T2/S2 (16200, 7200) code.