CONTROLLING WHEELCHAIR BY ANALYSING BRAIN NEURAL AND HEAD MOVEMENTS
Abstract : Brain Computer Interface (BCI) sometimes called Brain-Machine Interface (BMI). It is a direct communication pathway between the Brain’s Electrical Activity and external device. Computer based system that acquires brainsignals, analyzes the man dtranslatesthemintocommandsthatarerelayed to an output device to carry out into desired action.
We can measurethebrain waves using technique known as ELECTROENCEPHALOGRAPHY (EEG). Successfully use of P300B CI has also been reportedforpeoplewithdisabilities resulting fromstroke, spinalcord injury, cerebralpalsy, multiple sclerosis and otherdisorders. Therefore, theBCIsystemmaybe used to improvethe quality of life of such patients. Hence the development and Implementation of BCI system is complex and Time consuming. Brain ComputerInterface(BCI) offersasolutionto independent mobility for peoplewithmovingdifficulties. This paper proposes a BCI to smart control of a wheelchair
EEGsignalsacquisition is mainlydone byplacing electrodesonthescalp. 10-20 systemisthe most commonstandardusedto place electrode.Someusesfivechannelbi- polarelectrodeswhilesomeuses12Ag/Clelectrodes.SimplercommercialBCIwithone electrode and a reference electrode onear are also used. Signal are acquired by placing the electrode in particular location of the brain mostly using10-20placementsystem. Normally for ERD system it is done by imagining certain body partwhich is active response and for ERP or SSVEP it is done by presenting a stimulus which is reactive response. Sometimes both are used. Electrode is placed using 10-10 system which is much denserelectrode placement strategy. Here subject is presented with an arrow pointing to left,right ordown which isdisplayed for3secondsduring whichsubject are askedtoimaginelefthand,righthandorfeetmovementfor8seconds.64channeldevices to capture EEG signal the device is sample at 512 Hz with ahigh pass filter at 1z and subject were ask to execute three mentaltask left hand imagination movement, rest and wordsassociation.14electrodestocaptureEEGsignal,andsampleat256Hz.Numberof electrodes used varies from one research to another usually multiple electrode
Pre-processingEEGsignalisto removetheartifacts.Thereare manysources for artifacts duringthe experimentation, for example:eyemovement,musclemovement, and cardiogenicarti facts. In thiswork, two types of filters are applied.
Median filter is non-liner filter and the conceptbehind itis toremovethe backgroundnoiseby finding the median value for each channel and then subtract the data from the median value as medianfilter will remove the EMG which is related to muscle artifacts.
High pass filter with 0.16Hz cutoff frequency is used to remove the DC offset. Sometimes EEG signals works as batteries due to the interference between sensors and thebrainscalp, therefore, removingtheDC offset is needed.
Moreover, EEGsignalsare converted fromtime to frequencydomainusing periodogramcommand in MATLAB to calculate the Fast Fourier Transform (FFT) and normalize the output to find the power spectrum density (PCD)is used Converting the signal to frequency domain. K is the number of nearest neighbors.
|