论文部分内容阅读
针对盲信号分离需要判断分离结果与源信号是否一致的问题,基于统计独立变量函数仍然保持“统计独立”的性质,提出了独立分量分析(ICA)的输出分量与源信号的一致性判断方法.该方法通过计算混合信号及其差分值混合矩阵的相关矩阵,根据ICA各分量对应的最大相关系数来判断ICA各分量与源信号的一致性.模拟计算和实验结果表明:若差分前后混合矩阵的最大相关系数趋近于1,则ICA输出分量与对应的源信号一致;为保证分离的准确,差分前后混合矩阵的最大相关系数不应小于0.8.
For blind signal separation, whether the separation result is consistent with the source signal needs to be judged. Based on the fact that the statistical independent variable function still maintains “statistical independence”, a method of judging the consistency between the output component of ICA and the source signal is proposed. In this method, the correlation matrix between the mixed signal and the mixed matrix of the differential values is calculated, and the consistency between the ICA components and the source signal is judged according to the maximum correlation coefficient of the ICA components. The simulation results and the experimental results show that if the mixed matrix The maximum correlation coefficient approaches 1, then the ICA output components are consistent with the corresponding source signals. To ensure the accuracy of the separation, the maximum correlation coefficient of the mixed matrix before and after the difference should not be less than 0.8.