计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (1): 149-152.DOI: 10.3778/j.issn.1002-8331.2009.01.047

• 数据库、信号与信息处理 • 上一篇    下一篇

基于AR模型的脑-机接口问题研究

唐 艳,柳建新,邹 清   

  1. 中南大学 信息物理学院 生物医学研究所,长沙 410083
  • 收稿日期:2008-07-03 修回日期:2008-09-02 出版日期:2009-01-01 发布日期:2009-01-01
  • 通讯作者: 唐 艳

Research on brain-computer interface based on AR

TANG Yan,LIU Jian-xin,ZOU Qing   

  1. Institute of Biomedical Engineering,Central South University,Changsha 410083,China
  • Received:2008-07-03 Revised:2008-09-02 Online:2009-01-01 Published:2009-01-01
  • Contact: TANG Yan

摘要: 在脑一机接口的研究中分类识别技术占有重要地位。将脑电信号中事件去同步化/相同步化现象作为特征信息,深入讨论基于AR模型的自适应算法(AAR)和多变量参数AAR模型算法(MVAAR)在脑电信号特征提取中的应用。结合三种分类器,对这两种算法进行了比较,实验证明两种方法的实验效果都很好,但是MVAAR算法比AAR算法能够达到更高的分类正确率,其阶次一般选取也比较低,数据仿真吻合度高,具有更强的通用性。

关键词: 脑电信号, 脑-机接口, 自适应自回归模型, 多变量自适应自回归模型

Abstract: Identification and classification technology plays an important part in study of the BCI system.This paper presents the methods based on the AR model of adaptive algorithm and multi-variable AAR model algorithm to extract feature information which is related with ERD/ERS in EEG.The coefficients of AAR and MVAAR models respectively are combined with three kinds of classifiers to classify different tasks.From the results,we can see MVAAR algorithm made higher accuracy than AAR.MVAAR algorithm which orders are relatively lower.It realizes multi-channel data input and is more practical.

Key words: electroencephalogram(EEG), brain-computer interface(BCI), adaptive auto regression(AAR), multi-variable auto regression(MVAAR)