When Germany's Karlsruhe Institute of Technology (KIT) released a prototype sensor for allowing smartphones to smell, it didn't merely replicate what a human nose can do. Once deployed, the phone will be able to detect aromas far more precisely than human noses. What is the enterprise IT potential here? Quite a bit.
Let's start with a few dangers. When employees smell smoke, electrical burning or a gas leak, companies lose valuable time trying to find the source, especially if the smell is coming from behind a wall. That's mostly because human noses are fine — to varying degrees, changing from person to person — at detecting smells, but the nose can quickly become accustomed to the smell. A smell-equipped phone, however, could indicate that the smell's concentration at this instant is x parts per million and that the concentration increases when moving north and decreases when moving south. That alone could potentially save lives.
There are also retail applications. What if a quick wave of a smartphone over a fish or meat display could immediately detect which have gone bad or are mere hours away from doing so?
How can a smartphone smell? According to KIT, "the electronic nose only is a few centimeters in size. The nose consists of a sensor chip equipped with nanowires made of tin dioxide on many individual sensors. The chip calculates specific signal patterns from the resistance changes of the individual sensors. These depend on the molecules in ambient air, differ for the different scents and, hence, are characteristic and recognizable. If a specific pattern has been taught to the chip before, the sensor can identify the scent within seconds. To start the process, the researchers use a light-emitting diode that is integrated in the sensor housing and irradiates the nanowires with UV light. As a result, the initially very high electrical resistance of tin dioxide decreases, such that changes of resistance caused by molecules responsible for the smell and attached to the tin dioxide surface can be detected."
Clearly, this research has a long way to go before handset manufacturers will be open to including such a mechanism, but who nose? (Sorry. Couldn't resist.) After all, in the ultra-tight packed innards of today's smartphone designs, "a few centimeters" is hardly trivial.
Then there's the training involved, to program the associated software to recognize the aromas that will be valuable to businesses and consumers.
That all said, this is potentially a game-changer. Consider a restaurant owner who walks into her kitchen and detects with her human nose a foul-smelling problem. With the massive number of conflicting smells — and kitchen staff constantly moving ingredients all around the kitchen — it might take some time to track the smell down. A smelling-equipped smartphone might be a far better tracking agent if — and this is a big if — it has already been trained to know what to look for.
A "new" bad smell that is the result of someone experimenting with different spices might be beyond the phone's database. Also, in an environment as filled with aromas as a restaurant kitchen, how can the business owner really know which smell the device is tracking?
The screen would presumably have to display the list of all identified smells and allow the owner to select one to track. For example, it might recognize 59 smells and one of them could be "rotting shrimp." That might tell the owner all that she needed to know.
Then there is the security potential. Today, secured environments can detect heat (infrared energy), movement (breaking a light beam) and weight (sensors in the floor), among other attributes. But what about smell? A system could identify all of the smells in a room while it's confirmed as secure and then signal an alarm if the smells change, as would presumably happen when an intruder walked in with all of those intruder smells. The key would be to flag any material change in a room's smell, as opposed to searching for a specific scent, which an intruder could mask.
What about in a healthcare facility? There have been many stories about dogs detecting cancer or imminent deaths in patients, based on the dog's acute sense of smell. Could smartphones be trained to also detect these clues?
This is a classic case for an A.I. machine-learning application. Sensors placed next to patients could note any changes and what the changes smell like. Then after it has explored enough patients, it could extrapolate (based on eventual traditional medical testing and examinations) what those smells mean. It’s similar to efforts to predict when machinery will fail based on their sounds, another quintessential machine-learning application.
The biggest challenge for corporate developers working on coding homegrown mobile apps is to creatively leverage the many attributes of current smartphones. They have gotten quite good at leveraging geolocation, speed of movement, video analytics, sound detection and capture, etc. Aroma detection may just be the next development hurdle.