中国邮电高校学报(英文版) ›› 2017, Vol. 24 ›› Issue (3): 83-90.doi: 10.1016/S1005-8885(17)60215-2

• Others • 上一篇    下一篇

Recognition of motor imagery tasks for BCI using CSP and chaotic PSO twin SVM

李端1,张洪欣2,Muhammad Saad Khan1,米芳1   

  1. 1. 北京邮电大学
    2. 北京邮电大学电子工程学院
  • 收稿日期:2016-11-14 修回日期:2017-04-07 出版日期:2017-06-30 发布日期:2017-06-30
  • 通讯作者: 张洪欣 E-mail:hongxinzhang@263.net
  • 基金资助:
    基于多源信息融合的电路芯片旁路分析方法研究

Recognition of motor imagery tasks for BCI using CSP and chaotic PSO twin SVM

  • Received:2016-11-14 Revised:2017-04-07 Online:2017-06-30 Published:2017-06-30
  • Contact: Hong-Xin ZHANG E-mail:hongxinzhang@263.net
  • Supported by:
    Study of integrated circuit side-channel analysis based on Multi-information fusion

摘要: Accurate modeling and recognition of the brain activity patterns for reliable communication and interaction are still a challenging task for the motor imagery (MI) brain-computer interface (BCI) system. In this paper, we propose a common spatial pattern (CSP) and chaotic particle swarm optimization (CPSO) twin support vector machine (TWSVM) scheme for classification of MI electroencephalography (EEG). The self-adaptive artifact removal and CSP were used to obtain the most distinguishable features. To improve the recognition results, CPSO was employed to tune the hyper-parameters of the TWSVM classifier. The usefulness of the proposed method was evaluated using the BCI competition IV-IIa dataset. The experimental results showed that the mean recognition accuracy of our proposed method was increased by 5.35%, 4.33%, 0.78%, 1.45%, and 9.26% compared with the CPSO support vector machine (SVM), particle swarm optimization (PSO) TWSVM, linear discriminant analysis (LDA), back propagation (BP) and probabilistic neural network (PNN), respectively. Furthermore, it achieved a faster or comparable central processing unit (CPU) running time over the traditional SVM methods.

关键词: brain-computer interface, motor imagery, twin support vector machine, chaotic particle swarm optimization

Abstract: Accurate modeling and recognition of the brain activity patterns for reliable communication and interaction are still a challenging task for the motor imagery (MI) brain-computer interface (BCI) system. In this paper, we propose a common spatial pattern (CSP) and chaotic particle swarm optimization (CPSO) twin support vector machine (TWSVM) scheme for classification of MI electroencephalography (EEG). The self-adaptive artifact removal and CSP were used to obtain the most distinguishable features. To improve the recognition results, CPSO was employed to tune the hyper-parameters of the TWSVM classifier. The usefulness of the proposed method was evaluated using the BCI competition IV-IIa dataset. The experimental results showed that the mean recognition accuracy of our proposed method was increased by 5.35%, 4.33%, 0.78%, 1.45%, and 9.26% compared with the CPSO support vector machine (SVM), particle swarm optimization (PSO) TWSVM, linear discriminant analysis (LDA), back propagation (BP) and probabilistic neural network (PNN), respectively. Furthermore, it achieved a faster or comparable central processing unit (CPU) running time over the traditional SVM methods.

Key words: brain-computer interface, motor imagery, twin support vector machine, chaotic particle swarm optimization

中图分类号: