Smart materials such as lead zirconate titanate (PZT) have been widely used for generating and measuring guided waves in solid media. The guided waves are then used to detect local defects for structural health monitoring (SHM) applications. In this study, a self-sensing system, composed of self-sensing algorithms and a self-sensing circuit equivalent to a charge amplifier, is developed so that a single PZT wafer can be used for simultaneous actuation and sensing. First, a PZT wafer is modeled as a single capacitor and a voltage source, and a so-called scaling factor, defined as the ratio of the PZT capacitance to the capacitance of the feedback capacitor in the self-sensing circuit, is estimated by applying known waveforms to the PZT wafer. Then, the mechanical response of the PZT wafer coupled with the host structure's response is extracted from the measured PZT output voltage when an arbitrary excitation is applied to the same PZT wafer. While existing self-sensing techniques focus on vibration controls, the proposed self-sensing scheme attempts to improve the accuracy of extracted sensing signals in the time domain. The simplicity, adaptability and autonomous nature of the proposed self-sensing scheme make it attractive for continuous monitoring of structures in the field. The effectiveness of the proposed self-sensing scheme is investigated through numerical simulations and experiments on a cantilever beam.