Energy Harvesting Techniques for Autonomous WSNs/RFID with a Focus on RF Energy Harvesting

Supply circuits that harvest energy from surrounding ambient or dedicated sources have drawn much interest recently for providing a possibility of energy-autonomy to the wireless sensing devices. The objective of this thesis is to optimize the power transfer efficiency of the RF/microwave energy transducers in WSN/RFID applications. For this purpose, analysis on the power utilization of the wireless devices at different working states has been done, which implies a space of improving the power transfer efficiency by employing a novel design concept in the RF/microwave energy transducers. In order to observe a deep insight of the charge-pump based energy transducer, an analytical derivation has been implemented based on a compact I/V model for MOSFET working in strong inversion and subthreshold regions. The derivation provides a mathematical direction for the impact of the power consumption of the wireless device on the input impedance of the charge-pump rectifier, which acts as a core element in the energy transducer. With expressing the input impedance of the rectifier into a shunt connection of a resistor and a capacitor, as the load current consumption reduces the shunt resistance increases dramatically while the shunt capacitance holds a relatively constant value. This work proposes a methodology of employing an adaptively adjusted matching network between the rectifier and the antenna in order to optimize the power transfer efficiency according to the instant power consumption of the wireless devices on different working states. For read-only wireless devices with no embedded batteries, like RFID transponders, a tiny storage capacitor of pico-farad which can be charged-up to a certain voltage in microseconds is usually employed as a DC supplier. During the communication between reader and transponder, the reader radiates RF power continuously to supply the transponder. Extra power supply is required to adjust the matching network electrically for optimal power transfer, which raises a new challenge to the batteryless devices. A solution is proposed in this work that an auxiliary rectifier with a smaller constant load current consumption is employed to supply the feedback control circuitries. Besides, the abovementioned methodology is also applied in charging-up procedure of a wireless device which employs a supercapacitor as its charge storage. The charging-up procedure is extended to hours due to the huge volume of the capacitive storage, and the charging speed becomes a critical issue. During the charging-up, the output voltage of the recti- fier increases exponentially, while the charging current reduces exponentially. The input impedance derived for steady-state is not precisely applicable yet theoretically directive in this situation. A novel application of adaptively tunable matching network in transient process is implemented to accelerate the charging process of the wireless devices.