一种基于FPGA的脑电分类算法实现

通过对脑电信号特征的分析,利用小波变换的多尺度分析技术对脑电信号进行特征提取,进而使用主成分分析算法对特征进行降维,并对降维后的信号使用Fisher线性判别方法进行分类。最后,利用VerilogHDL硬件编程语言设计实现了Mallat分解算法、PCA算法和LDA算法模块,并在FPGA应用板上实现了脑电分类功能。系统对2008年BCI大赛的数据进行了测试,分类准确率达到92.31%,表明该方法对开发便携式脑机接口系统具有良好的应用价值。

EEG sorting algorithm based on FPGA

After analysis of the EEG signal characteristics,the EEG signal feature was extracted with multiresolution analysis technology of wavelet transform.The principal component analysis(PCA) algorithm is applied for dimensionality reduction of the features.Then,the signals after dimensionality reduction is sorted with Fisher linear discriminatory analysis(LDA) algorithm.Finally,the algorithm modules of Mallat decomposition,PCA and LDA are designed by using the hardware programming language Verilog HDL,and the function of the EEG recognition system is realized on the FPGA application board.The data obtained from the 2008 BCI contest was tested by the system.The classification accuracy reached 92.31%.The results show that this method has good application value in developping portable brain-computer interface system.