This paper presents results of verification of human subjects, using electroencephalography (EEG) signals taken from frontal electrodes. Although we performed examinations with a full EEG 10-20 montage system, we decided to assess the results obtained from different small sets of frontal electrodes to consider subset of electrodes which would have the performance in people verification system based on their EEG signals. Additionally, for reference, we included results achieved from all 19 electrodes. The sets of electrodes were chosen for their easy access and the possibility of using in commercially available EEG headsets or bundles. Over 700 examinations were collected from 36 healthy adults (almost 20 EEG sessions were recorded from each person). The first 15 examinations from each participant were used for training the feedforward neural network with the Levenberg-Marquardt backpropagation algorithm. For each person, the different neural network was created. In the next step, people were verified using these networks by means of statistical metrics evaluation. The metrics were calculated using examinations which were recorded on different days both for the training and testing sets. Thanks to this approach, we were able to exclude the impact of daily changes in EEG and to get closer to the actual use of such a system. In this preliminary study, we focused on spectral features of EEG signals, to investigate whether temporary changes would prevent people from being recognized. Due to the large dataset, this paper can be the introduction for further works on searching and selecting features that will allow verification and will be invariable in time.
KEYWORDS: Electroencephalography, Electronic filtering, Digital filtering, Linear filtering, Signal processing, Filtering (signal processing), Digital signal processing
This paper presents the evaluation of adaptive filtering methods for suppression of the second powerline harmonic in electroencephalography (EEG) signals. The powerline interference with its harmonics is the most frequent noise source distorting EEG signals. Mostly its power is too high, to be simply removed by a low-pass filter, especially during the analysis of upper gamma frequencies (up to 100 Hz), where some information about EEG signal could be lost. This paper focuses on comparison of the adaptive algorithms (Least Mean Squares (LMS), and Recursive Least Squares (RLS)) in the suppression of harmonic interferences. This evaluation is based on the dedicated measures, allowing to assess the distortions remaining after the powerline suppression. The results of studies confirm the usability of the adaptive filters in powerline and its harmonics suppression.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.