|
Distributed, self-powered, micro-scale, wireless sensor networks promise to do for large-scale systems what the integrated circuit (IC) did for computers and portable devices, changing the way they are studied, built, monitored, and controlled. In biomedical applications, encapsulated sensors can be implanted or swallowed to monitor various body functions and deliver medication on-demand. Industrial systems could distribute sensors throughout a plant or facility to accurately and uniformly control humidity, temperature, and countless of other parameters, and even perform system prognosis and initiate self-healing sequences. Sensors used in military vehicles can gather security-threatening information, such as the presence of toxic, explosive, or electromagnetic interference (EMI), to not only warn and automatically react but to also study the way systems behave once they are deployed in the field, allowing next-generation designers to improve the way the systems are built. Key enabling attributes to this technology are its high sensor-density and non-invasive features, which present challenges in the form of integration (including micro-scale energy sources); efficient power-conditioning microelectronics; and micro-Watt, load-managed transceivers, sensors, and supporting circuits, the composite of which is illustrated, in general form, in Figure 1.
Figure 1. Complete self-powered, wireless sensor node system
Integration
Integrating the energy source into a micro-chip is arguably the first challenge to confront. The device must pack sufficient energy into a small volume of space to yield operational lives on the order of days, months, and even years. To this end, thin-film lithium-ion batteries (Li Ion) are conformable and yield reasonable energy levels (that is, lifetimes), but not enough to sustain a practical electronic system when constrained to such dimensions, which is why micro-scale MEMS fuel cells are promising. Fuel cells, however, fall short in power, when compared to Li-Ion technology under similar space constraints. A hybrid micro-scale MEMS fuel cell-thin-film Li Ion source is therefore optimal [1]. Managing such a hybrid source requires multi-directional energy-flow and power-conditioning systems, as shown in Figure 2, wherein, as an example, a current-regulated boost converter is used to transfer energy from a 0.8 - 1.4V, two-stack fuel cell to a 2.7 - 4.2V Li Ion and a voltage-regulated buck (or boost) converter to transfer energy from the Li Ion to the wireless sensor load. The total energy and peak power the system can store and deliver is still limited to its overall dimensions, and a low power-consuming load is still required. Harvesting energy from the environment, be it kinetic, solar, and/or thermal, in situ can potentially increase the total energy of the system, but not the peak power available to the load, since a harvester of this scale can only garner pico-Joules at a time.
Figure 2. Hybrid micro-scale MEMS fuel cell-thin-film Li Ion source system
Integrating the antenna, sensors, and power inductors shown in Figures 1 and 2 also presents challenges. For one, the antenna's efficiency is closely linked to its geometry and driving carrier frequency: its size must be on the order of, or longer than, the carrier wavelength. Mainstream CMOS devices, however, provide gain up to approximately 1GHz, considerably below the optimal range of micro-scale antennas, which radiate efficiently at or above 50GHz. GaAs or SiGe technologies are optimal for maximum efficiency, but the mediocre efficiency levels (for example, less than maybe 10%) micro-scale antennas can produce when driven with CMOS-compatible carrier signals must be considered for cost reduction [3]. Similarly, high-quality inductors are useful in maximizing the efficiency of power-conditioning circuits, yet on-chip inductors have poor quality factors (Q is less than 5) and relatively low inductances (less than 100nH). Fortunately, state-of-the-art micro-Henry off-chip inductors with reasonable Qs now conform to 2mm x 2mm x 1mm dimensions, making co-packaged solutions and integration possible. Similarly, while on-chip piezoelectric and temperature sensors are possible, a wide-range of higher performance co-packagable pressure, piezoelectric, and humidity sensors also conform to micro-scale dimensions.
|
