To supply energy to implanted devices, wireless electromagnetic energy transfer is an effective and commercially proven technique. It involves transmitting electromagnetic energy into the body and collecting it via a coil or antenna. Low-frequency electromagnetic energy transfer is currently used in cochlear implants, radio-frequency identification (RFID) implants, retinal prosthetics, and neurostimulators (9, 20). The physical process underlying low-frequency wireless energy transfer is near-field electromagnetic induction. Near-field induction involves a transmitter coil generating a changing magnetic field, typically at carrier frequencies in ISM frequency bands such as 14 MHz and 27 MHz. The changing magnetic field induces an AC electrical current in a coil at the receiver. The AC signal is then rectified to DC and regulated to a stable voltage. Alternatively, rectification is not required if logic is operated directly off an AC supply (21).
Near-field wireless energy transfer results in electromagnetic fields that can heat tissue and generate electromagnetic interference on nearby electronics. Hence, the wireless link must meet regulatory guidelines, including those proposed by the FCC in the United States and the ETSI in Europe (22). For small implanted applications, where volume constraints limit the size of the receive coil, the practical range of energy transfer is limited to a few centimeters. To increase the transfer distance and increase efficiency, both transmitter and receiver can be tuned to induce resonant coupling (23). Alternatively, longer-distance communication can be achieved by operating at frequencies above hundreds of megahertz and employing far-field power transfer (19). However, often, these high frequencies cannot easily penetrate the skin, and channel losses in the body keep overall efficiencies low.
Energy gathering is most useful and flexible when harvesting ambient energy, in which the source of energy exists inherently as part of the system. This is not the case for near-field wireless energy transfer, where an explicit powered transmitter is required. An emerging approach that exploits true ambient energy is vibrational energy harvesting from the user's movement. Microscale vibrational energy harvesters have been shown to generate approximately 5 μW cm−3 from human motion (14). This has been applied to self-winding watches, which produce 5 μW on average and 1 mW when forcibly shaken. For higher-power applications, tens to hundreds of milliwatts can be harvested by scavenging energy from the heel strike of a person's gait (14).
For implanted applications, vibration-to-energy converters must be fabricated on the microscale (24). Microelectromechanical systems (MEMS) technology is ideal for these applications, as it allows for the fabrication of mechanical systems equal in size to a microchip. One example of MEMS technology is the MEMS energy harvester presented in Reference 16, which has been fabricated for a low-power biomedical system that includes a power controller and a programmable digital signal processor. The MEMS device operates as a transducer where mechanical motion results in a varying capacitance. Figure 2a presents the charge-constrained energy conversion cycle using the MEMS variable capacitor as part of a power converter. Charge is transferred to and from the variable capacitor when its capacitance is at a maximum and minimum, respectively. In Reference 16, this charge transfer must be synchronized; however, it is possible to reduce system complexity and power consumption with an asynchronous architecture (25). Figure 2b shows the fabricated MEMS prototype (26). The MEMS transducer implements a variable capacitor ranging from 2 pF to 260 pF and generates 5 μW of usable energy from a vibration source of 2520 Hz.