We tested a small-scale experiment that is dedicated to the study of the wave separation algorithm and to the velocity variations monitoring problem itself. It handles the case in which velocity variations at depth are hidden by near-surface velocity fluctuations. Using an acquisition system that combines an array of sources and an array of receivers, coupled with controlled velocity variations, we tested the ability of beam-forming techniques to track velocity variations separately for body waves and surface waves. After wave separation through double beam forming, the arrival time variations of the different waves were measured through the phase difference between the extracted wavelets. Finally, a method was tested to estimate near-surface velocity variations using surface waves or shallow reflection and compute a correction to isolate target velocity variations at depth.