We’ve had our fingers, voices and irises scanned, but there’s now a new biometric en vogue – ears.
NEC, the inventor of this new personal identification technology, says it has an accuracy rating of 99%.
It measures the unique effect your ears have on sound. By identifying how sound resonation is changed by the unique pattern of each person’s ears, security systems can now distinguish accurately between millions of individuals.
In case you’re wondering what the effect of modifications to your ear shape are, don’t worry. Those over sized ear rings and studs and the severe boxing ring pummelings you’ve imposed on yourself won’t affect the accuracy of the system. The new system works by measuring how sound is determined by the shape of human ear cavities to distinguish individuals.
The advantages of the new system are that it is more natural. It does not require particular actions, such as scanning a part of the body over an authentication device, which makes it easier to conduct continuous authentication, according to a statement from Shigeki Yamagata, general manager, Information and Media Processing Laboratories, NEC Corporation.
The system works everywhere, even when the user is moving and working.
For those not already sold on the idea, here’s the technical details of how it works. For a few hundred milliseconds, an earphone with a built-in microphone generates acoustic signals from the earphone speaker.
It then receives the signals transmitted within the ear through the microphone. During this process, the soundwaves transmitted are changed by the time they are received back. This varies from ear to ear. The data on the measurement of those changes created by each ear gives every person their unique digital signature.
The change measurement is made using a synchronous addition method, which adds and obtains the average of the waveforms of the multiple signals received. This is used to eliminate noise from the received signals. The system then calculates how the sound resonates within the ear – i.e. the acoustics of each ear.
All this happens within a second.
NEC tests have shown that there are two main sets of sound data that can be used for recognition. Firstly, there are the signal components that travel through the external ear canal and are reflected by the tympanic membrane. Secondly, there are signal components that pass through the tympanic membrane and are reflected within the inner parts of the ear.
NEC plans to commercialise the technology around 2018.
A wide range of applications is planned, including fraud and identity theft prevention. It will help to secure critical infrastructure and take the risk out of wireless communications and telephone calls, NEC says.
Image of Ear courtesy of Shutterstock.com
7 comments on “Forget fingerprints, ears are so next season in biometrics”
What does ear wax build-up and removal do to this thing?
Or things like colds, flu, or other rhinoviruses, or allergies, or ear infections, or even changes in barometric pressure? Seems like those factors would affect the size/shape of the ear canal (due to inflamation/swelling and/or the tension of the tympanic membrane.
99% leaves a lot of false positives/negatives in a large user base
What about. Hearing aids?
sorry, what’d you say?
As a secondary authentication method, this may have merit, although the only 99% accuracy and the effect of wax and other hearing-modifying ailments are a bit worrisome. Beyond all that, are we really saying this is less of an action than placing a fingertip on a button? Typically you’re already holding your phone, so it’s a minor inconvenience. Putting headphones in when you wouldn’t otherwise be wearing them is what makes this a less-than-ideal solution in my opinion.
I can’t wait until the music companies start licensing this to put DRM on everything. “Sorry, you can’t listen to this music because the aural signature of your ear canal does not match that of the person who purchased this track.”
What about hearing loss or deafness? What does it play into this measurement?