Cognitive Wireless Charger: Sensing-Based Real-Time Frequency Control For Near-Field Wireless Charging

A recent increase in mobile and IoT devices has led to the advancement of wireless charging. The state-of-the-art wireless charging systems operate at a particular frequency, controlled by the explicit networking from the power-receiving device (which relays the battery status information, useful for the frequency selection), but such control is not designed to cope with the variations in the power receiving device's placements and alignments (which are more significant in near-field and pseudo-tightly coupled charging applications, as more charging pads are being deployed in the public domains and serving heterogeneous clients). In this work, we analyze the impact of the power transfer performance caused by the power receiver's load, distance, and coil alignment/overlap and introduce cognitive wireless charger (CWC), which adaptively controls the operating frequency in real-time using implicit feedback from sensing for optimal operations. In addition to the theoretical and LTSpice-based simulation analysis, we build a prototype compatible to the Qi standard and analyze the performance of CWC with it. Through our analyses, we establish that frequency control achieves performance gains in inductive-coupling charging applications and is sensitive to the variations in the placement and alignment between the power-transmitting and the power-receiving coils. Our prototype, when CWC is turned off, has comparable performance to the commercial-grade Qi wireless chargers and, with CWC enabled, demonstrates significant improvement over modern wireless chargers.