论文部分内容阅读
最小和算法(MSA)折中了译码性能和运算复杂度两个方面,是低密度奇偶校验码(LDPC码)硬件实现最常用的译码算法。比特后验概率对数似然比(LLR)是LDPC码MSA译码的关键参数,现有的高阶调制信号比特后验概率LLR计算方法及简化算法都需要估计噪声方差,估计值影响译码性能。论文从分析M阶无记忆二维调制信号比特后验概率LLR通用的计算方法入手,研究了适用于MSA译码的高阶调制信号比特后验概率LLR简化算法,该算法无需估计噪声方差,进一步降低了运算量和实现复杂度。
The least-sum algorithm (MSA) compromises both decoding performance and computational complexity, making it the most commonly used decoding algorithm for hardware implementation of low-density parity-check codes (LDPC codes). The bit-wise posterior probability log-likelihood ratio (LLR) is a key parameter of LDPC code MSA decoding. The existing high-order modulation signal bit posterior probability LLR calculation method and the simplified algorithm all need to estimate the noise variance, and the estimation value affects decoding performance. In this paper, we begin with the general method of calculating the posterior probability LLR of M-level memoryless two-dimensional modulated signal bits, and study the LLR reduction algorithm of high-order modulated signal bit-posterior probability suitable for MSA decoding. Without estimating the noise variance, Reduce the amount of computation and implementation complexity.