In this paper, we present a fault diagnosis system for analog circuit testing based on the Simulation Before Test (SBT) approach. The circuit under test is subjected to a frequency response analysis, from which parameters are extracted such that they give unique values for every configuration of the circuit. This uniqueness makes them ideally suitable for utilization as signatures that may be used to characterize the state of the circuit. Software implementation has been done using MATLAB and SIMULINK, whereas, for the hardware implementation of the system, we have chosen the NI ELVIS platform. In both cases, signatures have been extracted from the frequency response of the circuit (Sallen-Key band pass filter). As most signature values obtained are unique, successful classification of faults is made feasible