This month we've got a carpet of plastic optical fibers that can be used to track walking patterns and thus provide early warning of potential falls, a hybrid metrology method for measuring truly tiny features on microchips, and a wireless sensor that spots whether a window is open or closed.
The Carpet That Maps Your Walk
Falls are very bad news if you're elderly. A fall a younger, healthy person would shrug off can send an older person to the hospital with broken bones. So, you ideally want to spot falls as soon as they happen, get an early warning of any increase in unsteadiness when a person walks (because the likelihood of a fall has just gone up), and generally monitor a person's walking patterns to figure out how they're doing health- and balance-wise. As described in the article, "'Magic carpet' could help prevent falls" the boffins at the University of Manchester, led by Dr. Patricia Scully, have developed a system that lets them map the footfalls of someone walking over a carpet. Below the carpet lie plastic optical fibers that flex as someone walks over them; on the edges of those fibers sit tiny electronics that relay the optical signals to a computer for processing. There, sophisticated software translates the optical signals into a real-time 2D image of the walker's footsteps. According to the article, the tomographic system could also be used to detect chemical spills. Either way, it's an elegant and unobtrusive system with valuable applications.
Measuring the Tiny
Moore's Law may give us better and faster computing, but the quest to cram more processing power into computer chips also creates a headache for those tasked with measuring whether the tiny features on the chips have the proper dimensions and thus work as intended. While direct measurement techniques (atomic force microscopy) and indirect optical techniques such as scatterometry can be used, the indirect methods aren't necessarily as accurate as desirable for the features as small as 22 nm and the direct method is pricey and slow. So, how do you get accurate measurements with low measurement error without breaking the bank? You use a hybrid metrology method developed at NIST that relies on a combination of measurement techniques, each of which measures different physical parameters, coupled with Bayesian statistical analysis to the reduce the resulting measurement error. The article, "NIST 'Hybrid Metrology' Method Could Improve Computer Chips" explains the rationale behind the hybrid metrology method and describes how the statistical analysis reduces the measurement uncertainty of the resulting data.
Window sensors tell you whether a window is open or closed; no surprises there. But what if you had access to a window sensor that provided that information to your phone? The Fraunhofer news article, "Wireless window sentinel" describes just such a device. The wireless window sensor that the article describes couples a triaxial magnetic field sensor in the inner frame of the window with a magnet embedded in the bottom of the sash. As the window is opened or closed, the sensor detects any changes in the position of the magnet and translates that into information on the state of the window. An RF node (also embedded in the window frame) relays the data to a base station and uses energy harvesting as its power source. And the base station can either display window status or provide window status in response to remote queries. (You know, the building managers at my old office would have loved these; our 'climate control' was such a joke that we were always opening windows so that we wouldn't freeze!)
Editor's note: I'm on vacation (joy!) next week, so there will be no Sensors Weekly on Friday, September 14!