电子与信息学报
電子與信息學報
전자여신식학보
JOURNAL OF ELECTRONICS & INFORMATION TECHNOLOGY
2015年
8期
1931-1936
,共6页
黄太奇%易本顺%姚渭箐%方华猛%李卫中
黃太奇%易本順%姚渭箐%方華猛%李衛中
황태기%역본순%요위정%방화맹%리위중
喷泉码%加性高斯白噪声信道%不等差错保护%扩展窗喷泉码%规则变量节点度
噴泉碼%加性高斯白譟聲信道%不等差錯保護%擴展窗噴泉碼%規則變量節點度
분천마%가성고사백조성신도%불등차착보호%확전창분천마%규칙변량절점도
Fountain codes%Additive White Gaussian Noise (AWGN) channel%Unequal Error Protection (UEP)%Expanding Window Fountain (EWF)%Regularize variable-node
该文提出了一种可适用于加性高斯白噪声(AWGN)信道的融合扩展窗喷泉码(Expanding Window Fountain, EWF)和规则变量节点度LT码(Regularized variable-node Luby Transform, RLT)策略的不等差错保护(UEP)算法,称为 EWF-RLT 编码算法。首先利用扩展窗口技术给不同重要等级的数据加窗,编码时让较高重要等级数据以更高的概率参与编码;同时,结合规则变量节点度算法,改变传统 LT 码编码过程中随机选取邻居节点的编码方式,使较高重要等级的数据具有较大的最小变量节点度,改善错误平层现象。分析和仿真结果表明,该文提出的EWF-RLT算法与传统算法相比,能对较高重要等级数据进行更强的保护,提升网络传输质量;在UEP方案设计中,加入RLT码编码参数,使得该文方案更加灵活与适用。
該文提齣瞭一種可適用于加性高斯白譟聲(AWGN)信道的融閤擴展窗噴泉碼(Expanding Window Fountain, EWF)和規則變量節點度LT碼(Regularized variable-node Luby Transform, RLT)策略的不等差錯保護(UEP)算法,稱為 EWF-RLT 編碼算法。首先利用擴展窗口技術給不同重要等級的數據加窗,編碼時讓較高重要等級數據以更高的概率參與編碼;同時,結閤規則變量節點度算法,改變傳統 LT 碼編碼過程中隨機選取鄰居節點的編碼方式,使較高重要等級的數據具有較大的最小變量節點度,改善錯誤平層現象。分析和倣真結果錶明,該文提齣的EWF-RLT算法與傳統算法相比,能對較高重要等級數據進行更彊的保護,提升網絡傳輸質量;在UEP方案設計中,加入RLT碼編碼參數,使得該文方案更加靈活與適用。
해문제출료일충가괄용우가성고사백조성(AWGN)신도적융합확전창분천마(Expanding Window Fountain, EWF)화규칙변량절점도LT마(Regularized variable-node Luby Transform, RLT)책략적불등차착보호(UEP)산법,칭위 EWF-RLT 편마산법。수선이용확전창구기술급불동중요등급적수거가창,편마시양교고중요등급수거이경고적개솔삼여편마;동시,결합규칙변량절점도산법,개변전통 LT 마편마과정중수궤선취린거절점적편마방식,사교고중요등급적수거구유교대적최소변량절점도,개선착오평층현상。분석화방진결과표명,해문제출적EWF-RLT산법여전통산법상비,능대교고중요등급수거진행경강적보호,제승망락전수질량;재UEP방안설계중,가입RLT마편마삼수,사득해문방안경가령활여괄용。
A novel scheme named EWF-RLT codes, which producesUnequal Error Protection (UEP) for Luby Transform (LT) codes over Additive White Gaussian Noise (AWGN) channel by using a windowing technique before regularizing variable-node distribution, is proposed in this paper. Firstly, the idea of “windowing” the data sets according to their protection requirements is applied to allow coded symbols to make more edge connections with more important parts of the information bit stream with high probability. Then, variable-node degree distribution is exploited to improve the error floor and ensure the more important class of information bit stream have a higher minimum variable-node degree by modifying the traditional method of choosing neighbor nodes randomly in encoding. Compared with the conventional UEP scheme, what is confirmed both theoretically and experimentally is that the proposed approach can provide significant performance improvement in themost important bits class and improve network transmission performance. Furthermore, the proposed scheme introduces additional parameters in the UEP LT code design, making it more general and flexible in terms ofthe realization of UEP scheme.