Human locomotion is based on complex interactions of several cortical and subcortical structures. Over the years, the main underlying mechanisms have been partially unveiled thanks to standard functional neuroimaging techniques as well as electroencephalography (EEG). However, a complete picture is still lacking to date, due to particularly challenging experimental difficulties arising on top of the inherent complexity of the involved mechanisms. In this context, the aim of this study was to investigate the EEG dynamics associated to the production of voluntary rhythmic foot movements only. We used an experimental protocol limiting drastically the presence of movement artifacts in the EEG signals compared to real walk on a treadmill. A time-frequency analysis was performed, based on a time-warping method allowing an ensemble averaging of the data of 3 subjects. Characteristic alternation of power increases and decreases in the alpha, beta and gamma bands during the movement cycle is demonstrated as well as the emergence of two different neural coordination schemes related to in-phase and anti-phase foot movements.