this paper presents a comparison among different time frequency representation methods in sleep study with EEG signal .EEG signal reflects brain activity and is useful for sleep study. Sleep study is necessary for diagnostic and treatment of sleep disorders.EEG is a non-stationary signal and therefore classic methods such as fourier transform is not suitable for studying it. Time frequency representation is one of the methods that are used for feature extraction of EEG signal. There are many approaches to time frequency representation and it is necessary to examine which of this approaches have better results for sleep classification. In this paper time frequency representation image is obtained from six time frequency …show more content…
EEG signal reflects brain activity and is useful for sleep study. Sleep study is necessary for diagnostic and treatment of sleep disorders
EEG is a non-stationary signal and a good method to studing it, is time frequency analysis. There are many approaches to time frequency representation and it is necessary to examine which of this approaches have better results for sleep classification. It is a Comparative analysis. Studies like this already existed. For example J.D. Mart´ınez-Vargas et al. do a Comparative analysis of Time Frequency Representations for discrimination of epileptic activity in EEG Signals.[1]
EEG signal has many useful features that can be used for sleep studying . Varun Bajaj and Ram Bilas Pachori proposed that features derived from the histogram of segmented time-frequency image (TFI) corresponding to frequency-bands of rhythms of EEG signals are useful for automatic classification of sleep stages[2] .For comparison among different time frequency distributions, it is necessary to have some features that should extracted from TFI. We use the features that Varun Bajaj and Ram Bilas Pachori proposed. These features will be introduced in methods and materials