在介绍 L DPC (L ow Density Parity Code)低密度校验码的基本原理的基础上 ,针对任意离散无记忆信道的传输 ,从两个方面对其结构进行了改进。这种改进的 L DPC码是定义在有限域 GF(q)上的非正则 L DPC码 ,较之正则 L DPC码具有更好的...在介绍 L DPC (L ow Density Parity Code)低密度校验码的基本原理的基础上 ,针对任意离散无记忆信道的传输 ,从两个方面对其结构进行了改进。这种改进的 L DPC码是定义在有限域 GF(q)上的非正则 L DPC码 ,较之正则 L DPC码具有更好的性能。采用改进的非正则 L DPC码 ,经过最大似然概率译码 ,能够实现以任意逼近任何离散无记忆信道容量的速率的可靠通信。同时 ,讨论了对应于这种码结构的实际的迭代译码方法 ,并简单介绍了这种改进的非正则 L DPC码在OFDM系统、压缩图像传输等方面的应用。展开更多
A parity check matrix construction method for constructing a low-density parity-check (LDPC) codes over GF(q) (q〉2) based on the modified progressive edge growth (PEG) algorithm is introduced. First, the nonz...A parity check matrix construction method for constructing a low-density parity-check (LDPC) codes over GF(q) (q〉2) based on the modified progressive edge growth (PEG) algorithm is introduced. First, the nonzero locations of the parity check matrix are selected using the PEG algorithm. Then the nonzero elements are defined by avoiding the definition of subcode. A proof is given to show the good minimum distance property of constructed GF(q)-LDPC codes. Simulations are also presented to illustrate the good error performance of the designed codes.展开更多
文摘在介绍 L DPC (L ow Density Parity Code)低密度校验码的基本原理的基础上 ,针对任意离散无记忆信道的传输 ,从两个方面对其结构进行了改进。这种改进的 L DPC码是定义在有限域 GF(q)上的非正则 L DPC码 ,较之正则 L DPC码具有更好的性能。采用改进的非正则 L DPC码 ,经过最大似然概率译码 ,能够实现以任意逼近任何离散无记忆信道容量的速率的可靠通信。同时 ,讨论了对应于这种码结构的实际的迭代译码方法 ,并简单介绍了这种改进的非正则 L DPC码在OFDM系统、压缩图像传输等方面的应用。
基金supported by the National Natural Science Foundation of China (60672087)
文摘A parity check matrix construction method for constructing a low-density parity-check (LDPC) codes over GF(q) (q〉2) based on the modified progressive edge growth (PEG) algorithm is introduced. First, the nonzero locations of the parity check matrix are selected using the PEG algorithm. Then the nonzero elements are defined by avoiding the definition of subcode. A proof is given to show the good minimum distance property of constructed GF(q)-LDPC codes. Simulations are also presented to illustrate the good error performance of the designed codes.