| 摘要: |
| 为了进一步提高欠定盲源分离算法中混合矩阵估计方法的性能，提出了一种基于加权最小二乘支持向量机（SVM）的欠定盲源分离混合矩阵估计方法。该方法利用信号的方向角度特征估计出有效信源信号个数，然后采用加权最小二乘支持向量机方法获得初始权值，每次将其中一个权值对应的样本点作为测试样本，其余点作为训练样本，依次对样本的误差变量进行更新，再根据权值计算公式实现所有权值的更新，进而确定最优分类平面，实现对观测信号的最优分类，最终估计出混合矩阵。实验结果表明，新算法是有效的，其平均误差是基于K-均值方法误差的0.2倍左右，是基于SVM算法平均误差的0.5倍左右。 |
| 关键词: 欠定盲源分离 加权最小二乘支持向量机 K-均值聚类 矩阵估计 |
| DOI： |
|
| 基金项目:国家自然科学基金资助项目（61271115）；吉林市科技发展项目（2013625009） |
|
| Undetermined blind source separation based on weighted least square support vector machine |
| ZHAO Liquan,LIU Shanshan |
| () |
| Abstract: |
| To further improve the performance of the underdetermined blind source separation algorithm，an algorithm based on weighted least square support vector machine(WLS-SVM) is proposed. Firstly,it estimates the number of source signal using the characteristic of frequency domain signal. Secondly,it uses WLS-SVM to obtain the initial weight values. The sample point corresponding to one of the weight values is used as the test sample every time,and the other is used as training sample. The error variable is updated sequentially,and then all weight values are updated according to the weight calculation formula to determine the optimal classification plane and realize optimal classification of observed signals to estimate the mixed matrix. Simulation results prove that the proposed algorithm has smaller error compared with traditional algorithm. The error of proposed algorithm is twenty percent of that of K-means based method,and fifty percent of that of SVM-based method. |
| Key words: undetermined blind source separation weighted least square support vector machine K-means clustering matrix estimation |
《电讯技术》编辑部 