Design of a new high-energy concentration kernel quadratic TFD for EEG spike signal

In this paper, the design of a novel high-energy concentration kernel quadratic TFD for EEG spike signal analysis is presented. Firstly, we show that the suppression of the negative frequency of the signal due of the use of Hilbert transform causes low Time-Frequency Distribution (TFD) resolution in the very low frequency band. To remedy this artifact, a frequency shifting of the signal to the mid frequency band is used so that the negative and positive frequencies are taken into account in the time-frequency domain. This process enhances the TFD resolution in the very low frequency band. Secondly, we derived a new separable kernel TFD with a high auto-terms energy concentration based on the localization of the auto-terms and cross-terms of the EEG spike signal in the ambiguity domain. The proposed kernel uses only two parameters and offers high TFD resolution compared to the existing ones.