论文部分内容阅读
讨论两个已得到重点研究的主题:调制样式识别和迭代接收机设计。文章将这两个主题融合于因子图架构中。因子图架构能够用一种统一的方法来表征一系列算法,尤其是迭代算法。特别地,文章描述了一种因子图,该因子图将调制样式识别融合于迭代接收机结构中。所提的迭代接收机通过因子图上的消息传递过程,寻找一条将调制样式识别、解调和译码结合在一起的最佳途径。进而构建了一个识别器,该识别器利用从译码器得到的反馈信息作为编码比特的一个先验概率。仿真结果证明所提的接收机相比于将调制样式识别跟解调译码分离实现的最大似然识别器在性能上有重大提高。
Discuss two key topics that have been highlighted: modulation pattern recognition and iterative receiver design. The article merges these two themes into a factor graph architecture. The graph structure can represent a series of algorithms in a unified way, especially the iterative algorithm. In particular, the article describes a factor map that incorporates modulation pattern recognition into an iterative receiver structure. The proposed iterative receiver through the factor map messaging process, looking for a modulation pattern recognition, demodulation and decoding the best way to combine. A recognizer is then constructed which uses the feedback information obtained from the decoder as a priori probability of the coded bits. The simulation results show that the proposed receiver has a significant performance improvement over the maximum likelihood discriminator realized by separating the modulation pattern from the demodulation code.