基于因子图的迭代信道估计与译码算法

Factor-Graph-Based Iterative Channel Estimation and Decoding Algorithm

  • 摘要: 针对频率平坦-时间选择性瑞利衰落信道下的数据检测问题,提出一种基于因子图与消息传递的联合迭代信道估计、符号检测与译码算法.对信息符号与信道系数的联合后验概率分布建立因子图模型,应用和积算法进行迭代消息传递,计算信息符号与信道系数的边缘概率分布.其中利用高斯参数化近似信道系数的连续概率密度函数,并结合前向-后向递归算法对信道系数进行迭代估计.仿真结果表明,在归一化多普勒频移分别为0.005和0.020的衰落信道下,该算法的误码性能与信道估计精度均优于传统的信道估计与译码算法.

     

    Abstract: Presented an iterative receiver for joint channel estimation, data detection and decoding over a time-varying frequency-flat Rayleigh fading channel. First, a factor graph (FG) that represents a joint posteriori probability mass function of the information bits and channel coefficients given the channel output was established. Then the sum product algorithm (SPA) was implemented on the FG in the proposed algorithm. Gaussian parameterization was employed to approximate the exact probability density function in message passing to yield a forward-backward recursion for the purpose of channel estimation. Compared with the traditional channel estimation and decoding methods, the effectiveness of the proposed iterative message passing algorithm is demonstrated through computer simulations.

     

/

返回文章
返回
Baidu
map