首页> 外文会议>Electrical amp; Electronics Engineering (EEESYM), 2012 IEEE Symposium on >Variable step-size blind source separation algorithm based on orthogonal gradient
【24h】

Variable step-size blind source separation algorithm based on orthogonal gradient

机译:基于正交梯度的变步长盲源分离算法

获取原文
获取原文并翻译 | 示例

摘要

Combined with natural gradient algorithm and FASTICA method, orthogonal algorithm was proposed for blind source separation. the algorithm by using the natural gradient algorithm iterative form and independent component analysis orthogonalization process, realizes the separation matrix in orthogonal space search. At the same time, the design of a new variable step size algorithm, which makes the step size in the iterative process of adaptive regulation, to further improve the performance of the algorithm. The simulation results prove that, variable step orthogonal gradient algorithm and the natural gradient algorithm and FASTICA algorithm, has better convergence speed and convergence performance.
机译:结合自然梯度算法和FASTICA方法,提出了一种正交算法进行盲源分离。该算法通过自然梯度算法的迭代形式和独立分量分析正交化过程,实现了正交空间搜索中的分离矩阵。同时,设计了一种新的可变步长算法,它使步长在迭代过程中进行自适应调节,从而进一步提高了算法的性能。仿真结果表明,可变步长正交梯度算法以及自然梯度算法和FASTICA算法具有较好的收敛速度和收敛性能。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号