An improved method to calculate phase locking value based on Hilbert–Huang transform and its application

详细信息    查看全文

• 作者：Jin Zhang (1) (3)
  Na Wang (2)
  Huan Kuang (3)
  Rulong Wang (3)
  
• 关键词：EEG ; Phase synchronization ; Phase locking value ; Hilbert–Huang transform
• 刊名：Neural Computing & Applications
• 出版年：2014
• 出版时间：January 2014
• 年：2014
• 卷：24
• 期：1
• 页码：125-132
• 全文大小：585 KB
• 作者单位：Jin Zhang (1) (3)
  Na Wang (2)
  Huan Kuang (3)
  Rulong Wang (3)
  
  1. College of Mathematics and Computer Science, Hunan Normal University, Changsha, 410081, China
  3. School of Software, Hunan University, Changsha, 410082, China
  2. Computer and Information Engineering College, Inner Mongolia Normal University, Hohhot, 010020, China
  
• ISSN：1433-3058

文摘

Phase synchronization analysis has been demonstrated to be a useful method to infer brain function and neural activity based on electroencephalography (EEG) signals. The phase locking value (PLV) is one of the most important tools for phase synchronization analysis. Although the traditional method (TM) to calculate PLV, which is based on the Hilbert transform, has been applied extensively, some of methodological problems of TM have not been solved. To address these problems, this paper proposes an improved method (IM) to calculate the PLV based on the Hilbert–Huang transform. For the IM, the Hilbert–Huang transform, instead of the Hilbert transform, is used to process non-stationary EEG signals and the empirical mode decomposition, not band-pass filter, is used to get target frequency band. The performance of the IM is evaluated by comparing normal and hypoxia EEG signals. The PLVs are used as features for a least squares support vector machine to recognize normal and hypoxia EEG. Experimental results show that the PLVs calculated by the IM can distinguish the EEG signals better than those calculated by TM.