Font Size: a A A

Application Research On P300-based BCI System

Posted on:2014-05-20Degree:MasterType:Thesis
Country:ChinaCandidate:C J YangFull Text:PDF
GTID:2268330392464393Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
A brain-computer interface (BCI) is a new kind of interface and communicationsystem that translates brain activity into commands for a computer or other devices. Inother words, a BCI allows users to act on their environment by using only brain activity,without using peripheral nerves and muscles. Therefore, to facilitate life for severelyparalyzed people also become the major goal of BCI research. In recent years, researchershave been interested in the study of BCI systems, in which P300-based BCI research is themost. This paper also does a lot of work in this regard.Firstly, we do a lot of research on P300potential. A P300-based online car system isrealized by using the characteristic of evoking without training. Based on the P300classicexperiment, we made some improvements, including changing the flashing charactermatrix into texts that represent certain sense, adding code to P300internal program sourcecode, and combining P300with wireless wheelchair prototype (car). BCI2000is used todo online analysis. Users can control the car without a large amount of training.Secondly, we expand the P300-based car system, and design a smart home controlsystem. The system uses BCI2000to do online analysis of the EEG, and then generatescontrol commands. Wireless receiving apparatus, which is designed with themicrocontroller and wireless module, is used for receiving control commands andcontrolling the state of smart home. The system inherits the advantages of P300,eliminates users the large amount of training. And it’s simple and reliable, easy to expand.At last, the CSP and CSP-L1method are deeply researched. CSP is based on theL2-norm which is sensitive to outliers because of the square operator. But the L1-normavoids the square problem. Compared with CSP, the new method combined L1-norm withCSP is more robust, and it is a good feature extraction method.
Keywords/Search Tags:Brain-Computer Interface, P300Potential, BCI2000, Smart Home, FeatureExtraction, Common Spatial Patterns, L1-norm
PDF Full Text Request
Related items