The prototype sensor promises to help address issues with current fingerprint scanners, which can be duped with fake or even similar fingerprints. The sensor was described in a paper - titled "A CMOS Proximity Capacitance Image Sensor with 16µm Pixel Pitch, 0.1aF Detection Accuracy and 60 Frames Per Second" - published in the Technical Digests of 2018 International Electron Devices Meeting.
"The most significant point of the developed sensor is its high capacitance sensitivity," says paper author Shigetoshi Sugawa, a professor in the Graduate School of Engineering at Tohoku University.
Many typical touch screen devices use a less sensitive capacitance sensor, where the differences in electrical properties between a sensor and a conductive tool - such as a finger - allow the device to react to scrolling or double clicking. The capacitance increases when the object is closer. The newly developed sensor, says Sugawa, derives its high sensitivity from newly introduced noise reduction technology.
The sensor chip contains pixels to detect the capacitance between the sample and detection electrodes. Each pixel has one detection electrode attached to it that's capacitively coupled with a ground wire. These electrical signals are converted into images of the samples. Previously, the signals would pick up background noise such as thermal noise and noise due to variability of electrical components of pixels, which made for lower-quality images.
The researchers addressed this by applying reset switches to the detection electrodes and employing a voltage pulse to produce a circuit that can follow the source of noise. The reset switches allow the systems to detect noise arising at the detection electrodes.
The voltage pulse alternates the two voltage levels after the reset switches are turned off, effectively canceling out and removing noise from the system - the equivalent, say the researchers, of removing the white and black snow from a television without signal input into a smooth, grey screen. It's much easier to sense any deviation on