用于稳态视觉诱发电位特征频率提取的同步压缩短时傅里叶变换方法

发布时间：2018-11-18 13:53

【摘要】：针对稳态视觉诱发电位(steady state visual evoked potential,SSVEP)范式下脑电信号(electroencephalograph,EEG)信噪比低、限制其识别正确率提高及脑-机接口应用等问题,根据EEG随机性、近似平稳的特点,提出了用于SSVEP特征频率提取的同步压缩短时傅里叶变换方法。该方法利用短时傅里叶变换对EEG进行时频分析,并通过同步压缩变换对时频平面的能量在频率方向进行重新分配,获得频率曲线更加集中的时频表达;同时为提高EEG信噪比,提取SSVEP脑电中特征频率附近信号进行重构,并利用典型相关分析进行分类识别,有效提高了最终识别正确率。仿真和实验结果表明,该方法极大地提高了信号的信噪比,具有良好的抗噪声性能和信号提取精度,且与传统的经验模态分解和常规滤波方法相比,该方法平均识别正确率最多分别提高了9.98%和4.38%,平均信息传输率最多分别提高了7.57bit/min和2.69bit/min,有效提高了SSVEP范式下脑-机接口的工作性能。
[Abstract]:In view of the problem of low signal-to-noise ratio (SNR) of EEG signal (electroencephalograph,EEG) under steady-state visual evoked potential (steady state visual evoked potential,SSVEP) paradigm, which limits the improvement of recognition accuracy and the application of brain-computer interface, according to the randomness of EEG, it is approximately stable. A method of synchronous compression short time Fourier transform (STFT) for SSVEP feature frequency extraction is proposed. In this method, the time-frequency analysis of EEG is carried out by using short-time Fourier transform, and the energy of time-frequency plane is reallocated in the direction of frequency by synchronous compression transformation, and a more concentrated time-frequency representation of the frequency curve is obtained. In order to improve the signal-to-noise ratio (SNR) of EEG, the signal near the characteristic frequency in SSVEP EEG is extracted for reconstruction, and the canonical correlation analysis is used for classification and recognition, which effectively improves the accuracy of the final recognition. Simulation and experimental results show that the proposed method greatly improves the signal-to-noise ratio (SNR) of the signal, and has good anti-noise performance and signal extraction accuracy, and is compared with the traditional empirical mode decomposition and conventional filtering methods. The average recognition accuracy is increased by 9.98% and 4.38% respectively, and the average information transmission rate is increased by 7.57bit/min and 2.69 bit / min, respectively, which can effectively improve the performance of BCI under SSVEP paradigm.
【作者单位】：西安交通大学机械工程学院;西安交通大学现代设计及转子轴承系统教育部重点实验室;
【基金】：国家自然科学基金资助项目(51275388)
【分类号】：R318;TN911.7

【相似文献】