This paper describes the use of strain stiffening in fixed-fixed beam actuators to extend the tuning range of microelectromechanical-systems-based Fabry-Perot filters. The measured wavelength tuning range of 1.615-2.425 mum is the largest reported for such a filter. Curvature in the movable mirror was corrected using a low-power oxygen plasma to controllably alter the stress gradient in the mirror. After curvature correction, the linewidth of a filter was 52 nm, close to the theoretical minimum for our mirror design. As a proof of concept, a filter was bonded to a broadband infrared detector, realizing a wavelength-tunable infrared detector. All measured data have been compared to theoretical models of the optics and mechanics of the filters, with excellent agreement between theory and measurement demonstrated in all cases. Finally, the Young's modulus and stress of the actuator materials were extracted directly from the measured voltage-displacement curves, demonstrating a novel technique for material property measurement.
An occupancy estimation sensor system based on low-pixel count sensor arrays is proposed. We evaluate a system comprising a 4 × 16 thermal detector array. Machine learning classifiers are used to interpret the raw data from the detector array for deducing the number of occupants in the sensor's field of view. We observe that nominal classifiers provide more robust classification than numerical classifiers. Furthermore, entropy-based classification is applied for the first time for occupancy estimation and found to produce the lowest root mean-squared errors and highest correlation coefficients compared with previously trialled classifiers.
Estimation of soil organic carbon stocks requires bulk density (BD) measurements. Variability in BD contributes to carbon stock uncertainty, in turn affecting how large a change in stock can be observed over time or space. However, BD is difficult and time-consuming to measure, and sample collection is further complicated by extremely dry field conditions, coarse-textured soils, and high coarse-fragment content, which are common in southern Australia and other semi-arid and Mediterranean-type climates. Two alternatives to reduce BD sampling effort are to take fewer BD samples at a site (i.e. volumetric rings or clod), and to use more time-efficient methods (i.e. gamma–neutron density meter, NDM). We evaluate these options in the context of a soil carbon stock survey in agricultural land in the south-west of Australia. The BD values within a monitoring site measured with conventional and NDM methods were statistically different when assessed using large sample sizes; the measurements diverged where the coarse fraction volume was >20%. However, carbon stocks were equivalent, reflecting the much larger relative variability in carbon percentage, which contributed 84–99% of the uncertainty in carbon stocks compared with <5% from BD. Given the maximum variability measured, soil carbon stock changes in southern Australia should be monitored on a decadal scale.
The monolithic integration of a low-temperature microelectromechanical system (MEMS) and HgCdTe infrared detector technology has been implemented and characterized. The MEMS-based tunable optical filter, integrated with an infrared detector, selects narrow wavelength bands in the range from 1.6 to 2.5 /spl mu/m within the short-wavelength infrared (SWIR) region of the electromagnetic spectrum. The entire fabrication process is compatible with two-dimensional infrared focal plane array technology. The fabricated device consists of an HgCdTe SWIR photoconductor, two distributed Bragg mirrors formed of Ge-SiO-Ge, a sacrificial spacer layer within the cavity, which is then removed to leave an air gap, and a silicon nitride membrane for structural support. The tuning spectrum from fabricated MEMS filters on photoconductive detectors shows a wide tuning range, and high percentage transmission is achieved with a tuning voltage of only 7.5 V. The full-width at half-maximum ranged from 95 to 105 nm over a tuning range of 2.2-1.85 /spl mu/m.
1 January 1992 in IEEE Journal of Quantum Electronics