Font Size: a A A

Research And Implementation Of The EEG Analysis Algorithm Based On FPGA

Posted on:2012-07-13Degree:MasterType:Thesis
Country:ChinaCandidate:H L BianFull Text:PDF
GTID:2298330467971710Subject:Circuits and Systems
Abstract/Summary:PDF Full Text Request
Brain-computer interface, BCI, refers to the communication devices between brains and electronic equipment such as computers. BCI dose not depend on the normal output pathways of peripheral nerves and muscles. The essence of BCI is to predict thoughts and objectives of people and realize the communication between people and machines. BCI technology provide a new to people with dyspraxia to communicate and control the environment, and make them live more normal lives.This paper focuses on the key technology of BCI, signal pattern recognition technology. After reaching on the characteristic of EEG signal, this paper select wavelet transform algorithm for EEG signal feature extraction. Due to the high dimension of wavelet feature, PCA algorithm is applied for dimensionality reduction. Then, Linear Discriminant Analysis algorithm, LDA, is used to classfy feature vectors.After the in-depth study of wavelet transform algorithm, this paper first implement Mallat wavelet decomposition algorithm on MATLAB and Quartus II. Then, PCA algorithm and LDA algorithm are realized on MATLAB to obtain the mean of training samples, the optimal PC numbers and PCA projection matrix that corresponding to the highest recognition rate, and find the LDA projection direction and the discriminative threshold. Those data are stored in the ROM on the FPGA chip to test and implement PCA and LDA algorithms using FPGA. At last, Mallat wavelet decomposition algorithm and PCA+LDA algorithm are integrated in one system that is implemented based on FPGA.In this paper, Quartus II development environment is used to design the hardware system, and the hardware platform is ITETEK-EP2C35-A development board. The FPGA chip on the development board is EP2C35F484C8, using JTAG download mode. After system debug, the EEG recognition system is realized on ITETEK-EP2C35-A development board.The experiment result shows that the recognition rate of the EEG recognition system in this paper is92.31%, and proves that the system is stable and has high recognition rate and practicality.
Keywords/Search Tags:Brain-Computer Interface, Wavelet Transform, Principal Component Analysis, Linear Discriminant Analysis
PDF Full Text Request
Related items