FPGA Design and Implementation of Wavelet Coherence for EEG Signals
MetadataShow full item record
The EEG waveform provides millisecond resolution brain information that can be obtained from the scalp using metal electrodes. It has become an applicable measure for a wide range of brain functionalities (including higher cognition) due to its low cost, non-invasiveness and ease of access. An important EEG application uses an evoked form of these signals linked to an external stimulus. For this thesis, an EEG was acquired during presentation of an oddball task and recording the event related potential (ERP), in which the P300 component is the most important. It reflects the participant’s response to rare or occasional stimulus events. Extracting features from these non-stationary signals can be achieved with a time-frequency method such as the continuous wavelet transform (CWT) whereas examining the functional connectivity between a pair of brain channels, as a source of EEG, can be achieved with the wavelet coherence (WC). However, the real time processing of these two digital signal processing (DSP) algorithms, which imply a large number of computations, requires running them with minimal delay for use in real time biofeedback applications. To achieve the required speed of processing for real time EEG applications, the involvement of hardware computation is required. One of the well-known hardware platforms in the field of DSP is the Field Programmable Gate Array (FPGA). These devices allow digital implementation of a wide range of DSP algorithms with a high processing speed, in addition to their configurability and portability. The aim of this thesis was the FPGA design and implementation of WC for EEG signals.
Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
School of Engineering
Item Access Status
In order to comply with copyright the article in Appendix A have not been published here.
Electroencephalogram (EEG) signals
Field Programmable Gate Array (FPGA) design