单通道混合信号盲分离算法

介绍了单通道混合信号的概念及盲源分离的现状,对实时线性混叠盲分离方法展开研究,分类探讨基于变换域滤波、多参数联合估计、符号序列与信道参数联合估计以及多维映射的单通道混合信号盲分离方法,分析比较各类盲源分离方法的处理对象、前提条件和优缺点,并进行了总结.最后通过仿真实验分析了基于粒子滤波与编码辅助的单通道盲分离方法中粒子数目、编码方式、源信号幅度比和频差等对分离算法性能的影响.

Blind separation algorithm of single-channel mixed-signal

This paper introduces the concept of single-channel mixed-signal and the status of blind source separation. On this basis, the blind separation methods for real-time linear-aliasing signals are discussed. The single-channel blind separation methods based on transform domain filtering, multi-parameter joint estimation, the symbol sequence with the joint estimation of channel parameters and mixed-signal multi-dimensional mapping, are studied. The application objects, preconditions, advantages and disadvantages of these blind separation methods, are analyzed and summarized. The effects of the number of particles, coding methods, as well as the amplitude ratio and frequency difference on the performance of the single-channel blind source separation algorithm based on particle filter and coding assistance are studied by simulation.