The Road to In-Display Fingerprint Authentication

Oct 04, 2016

By Anthony Gioeli


One of the many advances being made in smartphones and tablets is the progression to a “whole-device” display. Some models already have full-width displays with little or no bezel or ink area on the sides. The next step will be a nearly full-height display; “nearly” that is, because while the home button will be gone, there will still need to be a front-facing camera.

The use of biometrics for user identity has become increasingly accepted and popular for both device unlock (user authentication) and transactions (authorization with user authentication). The reason is that biometrics are more convenient than passwords and PINs, and have other advantages that enhance the security of mobile devices.

On smartphones and tablets, fingerprints are by far the most common form of biometric identification. Indeed, a fingerprint sensor is now available on nearly all flagship and other high-end models, and is quickly becoming a standard feature on mainstream and even some value devices, as well, owing to the desire to provide better security with a more satisfactory user experience. In most models, regardless of the price, the fingerprint scanning function is usually performed by a dedicated sensor that is exposed on the home button on the front of the device, but can also be found on the backside.

While eliminating the home button might appear to favor moving the fingerprint sensor to the back or side of the device, further developments in this direction are being reconsidered as some China OEM studies revealed that such designs were deemed too awkward and even difficult to use – whether an unnatural location when holding, or inaccessible when laying on a desk. Although the engineering is not trivial, the current direction is to locate the fingerprint sensor under the cover glass, which is preferred based on advantages in industrial design that facilitate improving usability, aesthetics, durability and reliability.

Initially the sensor will be located in the bezel/ink area of the cover glass where the button was located. The next step is placing the sensor in the lighted display area, likely in a fixed location at the lower edge just above where the bezel is today. Such a design will make it possible for applications to call attention to the sensor by highlighting it somehow when authentication or authorization is required; for example, by displaying a prompt under the sensor. Following this step is the ultimate goal, which is to enable fingerprint scanning anywhere in the display.

Moving the fingerprint sensor under the glass is a major design change and will, therefore, require a next-generation technology that is expected to evolve in three distinct phases.

Phase 1 involves locating the sensor under the glass in the bezel/ink area, and will mostly leverage existing capacitive sensing technology. Capacitive sensing is currently the dominant technology by far when it comes to fingerprint sensors, as they are proven and cost-effective.

Capacitive sensing works by discerning minute changes in an electric field, and that requires obtaining a sufficient signal-to-noise ratio (SNR). The signal in this case is the detection of the many tiny “ridges” and “valleys” in the fingerprint. The electrical noise comes from the display itself, and is quite “loud” by comparison.

Achieving a sufficient SNR requires the finger to be relatively close to the sensor, and the state-of-the-art today is a distance of 0.35mm or 350 microns. Because cover glass is thicker than this, it is necessary to shave or thin the topside or underside of the glass where the sensor is mounted, or use glass buttons.

Another possible technology involves acoustic scanning in the ultrasound range. Ultrasonic scanning can be made quite sensitive, but the high power consumption and/or high cost will likely make this technology viable only for specialty devices.

Optical technology is another candidate as it is not limited to the 350 microns range that limits capacitive technology. Note that optical needs to account for the lack of transparency in the bezel and has to provide a light source.

Phase 2 will involve moving the sensor to a fixed location within the display area. The extremely low SNR with capacitive sensing for glass thickness above 0.35mm will, therefore, render this technology impractical. With the sensor now able to function through clear glass, optical technology becomes increasingly attractive and even preferable because the photo diode sensors will be able to take advantage of the display itself as the requisite light source.

Phase 3 presents the most difficult challenge: enabling fingerprint scanning anywhere within the display. Just as capacitive sensing has become the technology of choice in Phase 1, the optical technology used in Phase 2 will likely become the preferred alternative in this final phase. For this reason, engineers can be expected to consider the challenges involved in whole-display scanning during Phase 2. Previous advances in touch/display integration will also have an influence in both of these latter two phases.

Phase 3 is the ultimate goal of fingerprint scanning on mobile devices. Full in-cell integration of the display with touch navigation and fingerprint scanning will enable persistent authentication that is performed continuously in the background, and in a way that is totally transparent and unobtrusive to the user. Devices will be wholly inaccessible to anyone but the owner, who will be able to conduct any and all interactions and transactions with utter ease and complete confidence. And the smartphone will become an even more indispensable part of our daily lives.