Gear elements are presented in a huge industrial and engineering application as well as in multiscale form like the actuator gear and the wind turbine transmission. They can work under different severe conditions. The gear transmissions are subjected to various kinds of excitations which can cause failure and mal function that can possibly lead to catastrophic scenario. In this work, a signal processing tool for damage detection of Wind turbine gearbox is investigated. In fact, the vibration responses generated by the developed numerical of the gearbox transmission are proceeded and treated using Discrete Random Separation (DRS) method. The obtained results show the ability of the proposed method to detect failure localized on the sun gear component. The method is simple to apply and does not require several parameters. It also shows a good capability to remove harmonics from the vibration response signal.