| Summary: | This thesis studies the phenomenon of sleep spindles. Sleep spindles are a distinct pattern that appears in electroencephalography signals during the second non-REM phase of sleep and they have been linked with brain functions like learning, memory consolidation and neuroplasticity.
An automatic method for spindle detection is implemented using an adaptive algorithm based on the spindles local average and maximum power. The method is also evaluated using an online database. After the detection occurs, the system further processes the spindle by converting it into a vector of parameters encapsulating information on the time domain, frequency domain, time-frequency domain and non-linear characteristics of the sleep spindle. Based on the above model, an embedded system was implemented using the Zedboard development board in order to verify the real-time performance.
|
|---|
