摘要
为了解决传统卢比变换(Luby Transform,LT)码在加性高斯白噪声(Additive White Gaussian Noise,AWGN)信道中存在高误码平台的问题,提出了一种具有逆向边增长(Reverse Edge Growth,REG)结构的LT码.REG-LT码会标记度数值相对较小的信息节点,并使其逆向精准地选取足够数量的校验节点,从而获得不低于期望目标的度数值.首先,设计了REG-LT码的编码算法,分析了校验节点和信息节点的选取原则.其次,推导了REG-LT码的信息节点度分布和校验节点度分布的数学表达式,分析了REG-LT码的收敛性.最后,引入增益损失比(Gain Loss Ratio,GLR)以衡量参数T;对收敛性的影响,将最大化GLR作为求解最优T;值的设计原则.仿真结果表明,与传统LT码相比,本文的REG-LT码可将误码平台降低3至5个数量级,且至少可获得2 dB的编码增益;此外,REG-LT码也能够实现优于现有LT码的误比特率(Bit Error Rate,BER)性能.
To solve the problem of high error floor for the conventional Luby transform(LT)codes in additive white gaussian noise(AWGN)channels,the LT code with reverse edge growth(REG)framework is proposed.For the REG-LT code,those information nodes with relatively small degrees will be marked,and these nodes will accurately select enough check nodes in reverse to obtain a degree that is not lower than the desired target.The coding scheme of the REG-LT code is designed,and the selection principle of check node and information node is analyzed.The mathematical expressions of the information degree distribution and check degree distribution of the REG-LT code are derived,and the convergence behavior of the REG-LT code is analyzed.The parameter gain loss ratio(GLR)is designed to measure the impact of the parameter T;on convergence behavior,and maximize the GLR is used as a design principle to solve the optimal T;.The simulation results show that compared with the conventional LT code,the REG-LT code in this paper can reduce the error platform by 3 to 5 orders of magnitude,and at least 2dB of coding gain can be obtained.Besides,the designed LT code outperforms the existing LT codes in terms of bit error rate(BER)performance.
作者
宋鑫
程乃平
廖育荣
倪淑燕
雷拓峰
SONG Xin;CHENG Nai-ping;LIAO Yu-rong;NI Shu-yan;LEI Tuo-feng(Department of Graduate Management,Space Engineering University,Beijing 101416,China;Department of Electronic and Optical Engineering,Space Engineering University,Beijing 101416,China)
出处
《电子学报》
EI
CAS
CSCD
北大核心
2022年第5期1201-1210,共10页
Acta Electronica Sinica
基金
国家自然科学青年基金(No.61805283)
军内科研项目(No.1700050400)。
关键词
喷泉码
LT码
高斯信道
误比特率
度分布
编码算法
外信息传递图
fountain codes
LT codes
Gaussian channel
bit error rate
degree distribution
coding algorithm
extrinsic information transfer chart
