Removing Artifacts And Extracting Patterns In EEG Based On ICA

ICA is a multi-dimensional signal-processing system recently developed technically with it's non-peak signal as its processing object.Under certain satisfied conditions,it can detach certain independent source of signals from the multi-circuit observation signals.ICA on its early stage primarily deals with blind separation issues similar to cocktail.In the 1990's,J.Herault and C.Jutten put forward the the basic conception of ICA from the research fo BBS.L.Tong analyses the detachability and uncertainty of the Blind Separation ,and developed an advance statistical retangular Algebra character analysis.Afterwards Comon gave a systematic interpretation to ICA.Based on the advanced statistics ,he directly framed the object sines.In 1995 A.J.Bell and J.Sejnowski reinterpret the Blind Separation issue from the perspective of information system ,and put forward the randomly graded decrease of Infomax ICA ,the starting point of the ICA study peak .Hereafter, S.Amari with his research group did a lot of reative work in the calculation theory study.In the following ICA research system some young scholars, such as T.W.Hyvarinen did outstanding contributionsin the field,who put forward the extended Infomax,FastICA calculation,making the ICA technology further into the field of application.Signals usually signals from different parts. The checking and recording equipment actually is a kind of signal retraction of different sources ,usually accompanied by noise disturbance source, which calls for the separation of signal source and the noise disturbance source using the ICA method, for the purpose of the brain electrical signal's noise disposal and character abstraction ,hereafter helping our further study of brain cognition .ICA's application to EEG signal processing analysis diaposal is a good instance in the field of ICA biomedicine signal processing.Furthermore,ICA has attracted great attention in the fields as communication ,model recognition and radar signal processing.The EEG disposal process includes other bio-electro signals {for example eyeimpulse}as well as outside disturbance signals.The purpose of brain electro signal disposal is to separate the useful brain electro signals from the complex background noise, and abstract some definite brain electro character in the sense of biology,and apply it to the research of medicine and brain cognition.The article gives a brief and consice introduction to the history and development of the Independent Quotient analysis and the lab brain electro signal collecting system and its application software system.The article gives a systematic analysis of the collected data according to the collected brain electro data,using the basic principles of ICA and its representative bicalculation,which will be used in the brain electro noise disposal and character abstraction study.The work is as follows:1. Based on the principle of Infomax transmission, FastICA algorithm and the basic principle of Infomax have been discussed, respectively. System analysis on the algorithm as well as their blind source separation performance has been conductedand confirmed. Obtained result has a certain guidance to the Infomax algorithm improvement and practical application.2. Studying the ICA-based method to remove EEG artifacts in order to efficiently extract ERP signal from the strong background noise. In the experiments we analyse the EEG independent components separated by ICA algorithm, by means of the time analysis, frequency analysis and scalp topography analysis, to find the noise components, this multiangle analytical method enhances the reliability of the components analysis. Moreover, the traditional ERP denoising based on ICA is often only used to remove non-neural artifacts such as EOG, besides that, we also remove the time-lock spontaneous EEG such as waves and μwaves to obtain better ERP denoising effect. The experimental results show, by the new ERP denoising method, ERP signal is efficiently extracted from the strong background noise.3. ICA is used to analyze the induced brain electrical signals. The experimental result indicates that ICA can clearly separate the brain electrical characteristics with a certain physiological significance. It is of great importance to the research on clinical medicine.