{"title":"A new reliability-based incremental redundancy hybrid ARQ scheme using LDPC codes","authors":"Hamid Saber, I. Marsland","doi":"10.1109/CWIT.2015.7255170","DOIUrl":null,"url":null,"abstract":"We present a new reliability-based hybrid automatic repeat request (RB-HARQ) scheme based on low density parity check (LDPC) codes. With the proposed RB-HARQ, which uses a rate-compatible LDPC code with puncturing and extending, the longest codeword is divided into clusters of code bits. Unlike previous works, in the event of a decoding failure, the receiver measures the reliability of received clusters, instead of code bits, and determines which cluster would be most beneficial for retransmission. Several metrics to determine the best cluster candidates for retransmission are derived analytically. We show that one of the new metrics outperforms the previous metrics. We also show that even with the feedback overhead taken into account, our RB-HARQ can still result in significant gain over the previous works, provided that the cluster size is appropriately selected.","PeriodicalId":426245,"journal":{"name":"2015 IEEE 14th Canadian Workshop on Information Theory (CWIT)","volume":"8 8","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 IEEE 14th Canadian Workshop on Information Theory (CWIT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CWIT.2015.7255170","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
We present a new reliability-based hybrid automatic repeat request (RB-HARQ) scheme based on low density parity check (LDPC) codes. With the proposed RB-HARQ, which uses a rate-compatible LDPC code with puncturing and extending, the longest codeword is divided into clusters of code bits. Unlike previous works, in the event of a decoding failure, the receiver measures the reliability of received clusters, instead of code bits, and determines which cluster would be most beneficial for retransmission. Several metrics to determine the best cluster candidates for retransmission are derived analytically. We show that one of the new metrics outperforms the previous metrics. We also show that even with the feedback overhead taken into account, our RB-HARQ can still result in significant gain over the previous works, provided that the cluster size is appropriately selected.