This paper deals with the wavelet analysis method for seizure detection in EEG time series and coherence estimation. The main part of the paper presents the basic principles of signal decomposition in connection with the EEG frequency bands. Wavelet analysis method has been used for detection of seizure onset. The wavelet filtered signal is used for the computation of spectral power ratio. The results show that our method can identify pre seizure, seizure, post seizure and non seizure phases. When dealing with seizure detection and prediction problems it is important to identify seizure precursor dynamics and necessary to identify information about onset and spread of seizures. Therefore in the second part the coherence and phase synchrony during pre seizure, seizure, post seizure and non seizure are computed. We expect this method to provide more insight into dynamic aspects of the seizure generating process.