Dynamic behavior and system identification are important topic in monitoring and maintaining existing infrastructures. System identification using Frequency Domain Decomposition (FDD) is an operational modal analysis (OMA) in frequency domain used on experiment of shear frame model with random vibration method is validated by comparing output of FDD data using acceleration input from simulated model with output of FDD using acceleration result of experimental model. The result of acceleration data is recorded using USB accelerometer X16-1D then calibrated and analyzed using Matlab programs, data_procces.m and solve FDD_eksperiment.m to estimate the modal parameter of model structure. Compared with parameter modal of simulation model, FDD method with input simulated acceleration resulted in difference of 1.757% in first frequency and 0.462% in second frequency. Meanwhile for FDD method using acceleration of experimental model, resulted in difference of 6.3126% in first frequency and 7.7327% in second frequency. FFD method is fairly accurate in predicting the frequency of structure, but for difference of mode shapes in experimental is very big compared to simulated model therefore it can be concluded that, this modal parameter is cannot be detected in experimental model.