There are a significant number of uncertainties in finite element models of unreinforced masonry structures related to the modelling assumptions and the properties of local materials. Therefore, it is necessary to implement calibration techniques for these models. Modal testing is a good option for assessing the dynamic properties of the structure. The experimental data is used to verify and improve the predicted response obtained by finite element model. The study presents the modal testing of a full-scale physical model of an unreinforced masonry house. The structure is tested under three different excitations: an impact by a calibrated hammer, a random excitation induced by a calibrated hammer, and a steppedsine excitation induced by a shaker. In addition, an operational modal test has been performed using ambient and random excitations. Two different methods are used for system identification: peak picking and stochastic subspace identification. The results of this research will be used in future studies for updating the model.