This article deals with the diagnosis of faults in rotor current sensors in wind systems based on the Doubly Fed Induction Generator (DFIG). An approach is developed to monitor rotor currents and identify faulty sensors, even during variable wind speeds. This method relies on the use of the Extended Kalman Filter (EKF) observer, which is based on the nonlinear model of the DFIG and a fault identification block in the abc reference frame. Firstly, the EKF estimates the rotor currents from stator current and mechanical speed measurements. Then, these estimates are processed in the localization block to precisely identify faulty sensors, whether it be single or multiple faults. This method has been implemented on a FPGA and has been experimentally validated. The experimental results confirm the effectiveness of this approach in monitoring rotor currents and identifying faulty sensors, in the case of single or multiple faults.