Using the human body as a unique link between gadgets will not lead to novel biometric modalities.
Recently, a couple of different groups have created prototypes that use the human body as a link between two gadgets, one mobile and the other, stationary.
The first used an acoustic signal transmitted from a smartphone through the user's body to a doorknob to unlock the door.
The second used electrical signals to transmit an MP3 file through the users body to a speaker system. That's pretty cool.
In their most basic use cases (using the body as a wire), these innovations accomplish little that couldn't be accomplished with a USB cable. But if these technologies come to incorporate a biometrics and ID management element, they could kick start a revolution in mobile computing and ID management.
It's not hard to see how future iterations of similar systems might use biometric modalities already in use — such as integrating a fingerprint reader with the conduction sensor for authenticating a data link — but both sets of innovators have something more profound in mind: using the electrical/acoustic properties of the body itself as an identifier.
The company is looking at different applications. Bhikshesvaran said the company was exploring the notion that it could end up being a new biometric footprint, since bodies all possess a unique energy signature. The company hasn't quite figured that one out yet.
and
Amento and his colleagues think they can add another layer of security to the smartphone key, too — one that's based on the unique properties of people's skeletons. Because of differences in bone lengths and density, people's skeletons should carry vibrations differently, they think.
My guess is that the fingerprint verification at one end of the link will be relatively straightforward, provide strong authentication and will work well enough to render the development of the new conduction/acoustic modalities impractical even over the very long term.
This is because in order to displace the well-understood modality of fingerprints and in order to make developing them worthwhile, the novel approaches will have to prove themselves to offer advantages far in excess of fingerprints (in order to justify the R&D outlay) and I don't see this happening.
Q: Are the electronic and acoustic properties of individuals stable?
A: Compared to fingerprints, I doubt it. Changing the chemistry or mass of a body will lead to minute changes in its electric or acoustic properties. Drinking a sports drink will change electrolyte levels and cause a tiny change in electric properties. Visiting a buffet, wearing a heavy backpack or changing shoes will change the acoustic properties of a person at least a little.
Q: Are the electronic and acoustic properties of individuals unique?
A: Compared to fingerprints, I doubt it. Fingerprints can be as funky as they want to be without killing anyone; not so with the chemistry behind conductivity or the skeletal structure of a person.
Q: How easy is it to measure the properties involved?
A: Conductive and acoustic properties may be unique enough for a team of doctors with infinite resources and lots of time to make a positive ID but not unique enough to enable a very fast, cheap and confident identification.
But the biggest reason these novel approaches are extremely unlikely to be adopted in the competitive marketplace is the very nature of the technology (skin on hardware) lends itself perfectly to the cheap, well-understood and reliable fingerprint tech. No other modality actual or theoretical stands to recommend itself more highly than finger/hand based biometrics and no profit seeking organization will likely devote the resources necessary to establish the reliability conduction/acoustic biometrics that will at best only ever be equal to fingerprints.
The more novel approaches will probably only ever be used as a method of weak authentication such as liveness testing so as to thwart the old rubber finger trick.
