Signal processing has essential role in features extraction of the system. In many cases, signals that must be processed, show periodic properties and they can be a sinusoidal signal model or a combination of sinusoidal signals model with time-varying parameter (amplitude, frequency and phase). Kalman filter, phase locked loop, adaptive notch filter and second order generalized integrator with frequency estimate loop are methods that can estimate parameters of sinusoidal signals. These methods mainly focused on sinusoidal signals with constant amplitude. However, in some practical applications such as the propagation of acoustic waves and the electromechanical oscillations in power systems, disturbances of signal amplitude changes fast and damping. Electromechanical oscillations in power systems caused by the interaction of different parts of system. Accurate estimation and characterization of these oscillations can increase situational awareness of the system and can even be used to take appropriate control actions to maintain the balance of the system and can help to prevent instabilities. In such applications, signals can be modeled with damped sinusoidal signals. These signal parameters include amplitude (primary and instantaneous), frequency, phase (primary and instantaneous) and damping factor which should be estimated in order to identification of signal. Most methods for estimating parameters of sinusoidal signals can be extended and modified in order to estimate the parameters of the damped sinusoidal signals. Recently Kalman filter and second order generalized integrator have been extended for this application. In this study, enhanced phase locked loop is modified in order to estimate the parameters of the damped sinusoidal signals. Dynamical equations of proposed method is obtained in two ways: 1) Based on the equivalent (in terms of mathematical formulation) enhanced phase locked loop and second order generalized integrator, and 2) using a gradient descent method and minimization cost function. These two methods are changed and convert to modified enhanced phase locked loop to estimate the parameters of the damped sinusoidal signals. Stability analysis shows that the values of the parameters of a damped sinusoidal signal (instantaneous amplitude, frequency, phase and instantaneous damping factor) are an asymptotically stable equilibrium point for the proposed method of dynamic system. Computer simulations confirm desirable performance of the proposed method to estimate the parameters of damped sinusoidal signals. Also, according to simulations, the proposed method as well as enhanced phase locked loop has good performance in very low sampling rate and robust operation in high sampling frequency. The proposed method can be used to extracted electromechanical oscillations on phasor measurement data that provided by the phasor measurement units across the power system. Key Words Electromechanical oscillations , phasor measurement units, damped sinusoidal signals, enhanced phase locked loop, frequency estimation , damping estimation .