Accelerometers are widely used in industrial applications and consumer electronics. We can find them in automotive crash detection or fitness trackers. The majority are based on piezoresistive or capacitive effect which are limited by their large size. This negatively influences their utility in emerging applications such as Internet of Things and biomedical applications. NEMS (Nanoelectromechanical systems) are presented as a possible solution for this problem. The accelerometers used in this project are made of a suspended graphene membrane and an attached silicon proof mass. They were fabricated by KTH researchers. We study the possibility of measuring accelerations by monitoring the changes in the device resonant frequency. If proven, to the best of our knowledge, they would be the first ultra-miniaturized and sensitive resonant accelerometer. To do so, we built a functioning model of the accelerometers by combining a theoretical framework and finite element simulations. The chip was placed on a piezo-shaker and measurements were conducted using a Laser Doppler Vibrometer. To complement the study, further measures were taken with an Atomic Force Microscope and a Digital Holographic Microscope. The results show a good match between theory and simulations. However, the acceleration measurements in the device show higher signals not related with accelerations. We proved they were related to displacements and theorized that they were caused by internal chip forces. Horizontal forces on the accelerometer frame could eclipse the effect of the accelerations on the resonant frequency. We suggest the use of a more rigid chip frame as a potential solution.