基于混合聚类和网格密度的欠定盲矩阵估计

欠定盲矩阵估计是欠定盲源分离的关键技术，其估计结果直接影响源信号的分离精度。针对目前欠定盲矩阵估计算法稳定性差、估计精度不高的缺点，提出了一种基于混合聚类和网格密度的新算法。该算法利用基于人工蜂群算法和K-均值的混合聚类方法对信号数据进行聚类，提高聚类结果的稳定性;利用网格密度法修正每一类的聚类中心，提高混合矩阵的估计精度。实验结果表明，所提算法在稳定性和估计精度方面都比传统欠定盲矩阵估计算法有了明显改善。

Underdetermined blind mixing matrix estimation algorithm based on mixing clustering and mesh density

The underdetermined blind mixing matrix estimation algorithm plays a key role in underdetermined blind source separation.The estimation accuracy of the algorithm directly affects the source signal.With regards to the low stability and the poor estimation accuracy of the current algorithm,a new algorithm based on mixing clustering and mesh density is proposed.The new algorithm clusters the signal data,taking advantages of the clustering algorithm based on artificial bee colony and K-means algorithm to ensure the stability of the clustering results,and improves the estimation accuracy using the method of mesh density.Experiment results show that the new algorithm has the advantages of high stability and estimation accuracy compared with the traditional underdetermined blind mixing matrix estimation algorithm.