Abstract
Wireless sensor networks (WSNs) have been widely used in applications including health, automotive, smart buildings, and predictive maintenance of structures. In these applications, there are thousands of sensor nodes depending on batteries which life time is limited. Thus, harvesting ambient energy provides a feasible alternative to sustain sensor nodes without the maintenance of batteries. Many energy sources are available, such as thermal, radio, vibrational, and solar energy. Due to the high power density and the compatibility with integrated-circuit (IC) technology, piezoelectric vibration-to-electricity conversion has received much attention recently. Since the piezovoltage is ac-type, the first stage of interface circuit is usually an AC-to-DC converter (rectifier). In the second stage, a charger as an impedance adaptor is adopted. To realize maximum power transfer, the equivalent impedance of interface circuit should match the piezoelectric transducer impedance. For deriving the adaptive interface parameters, the maximum power point tracking (MPPT) technique is necessary. Traditional perturb and observe (P$\&$B) MPPT technique invests long tracking time and large power consumption that is unsuitable applying in low-power system. Therefore, we introduce a new technique reverse-charging to detect the transducer parameters and thus track the precise interface parameters. Simulations have shown that the overall efficiency is 82.3.