A parametric identification technique via closed-loop testing using frequency response techniques is presented in this study. Under an operation of proportional control, a single dynamic test (such as an arbitrary pulse or step change) in the set point was introduced to the system. The closed-loop transients could numerically be translated into frequency response data by Fourier integral transforms, and the parameters of a given process model were then be obtained by a least-squares fit in the frequency domain based on Parseval＇s theorem. This identification technique could apply not only for self-regulating processes but also for integrating or open-loop unstable processes. Furthermore, the proposed technique was also extended to the controller settings for the feedforward-feedback control system. Simulation results demonstrated that the proposed technique could yield good model parameters and can be applied in many different types of processes.