Objects fitting the IoT (Internet of Things) definition present the challenge of balancing security with ease of use. Those who design and use them have been struggling to balance this equation.
A recent discovery by researchers at Northwestern Polytechnical University, has potentially solved this problem.
The IoT issues
The problem with having all of your electronic devices connected together and controlled over your home’s WiFi is that, despite the many benefits, it opens them up to attack by opportunistic hackers. Researchers and designers have been attempting to find security measures suitable for objects inside the home while maintaining their sease of use. Biometrics like fingerprint scanning are very secure, but are a pain to use every time you want to utilise a piece of tech within your home. It doesn’t sound like much, but unlocking every single device in your home with your finger repeatedly will reduce their lifespan.
The same goes for some other suggested methods that have been deemed workable but ultimately impractical. No-one is going to get a microchip in their skull so that only they have access to their dishwasher. On the other hand, leaving your internet-of-things unsecure is almost out of the question and would be a sure-fire way to get your data stolen. Just a single backdoor into your network can have disastrous results.
The Freesense Solution
This recent discovery by the team at Northwestern Polytechnical could be the answer. They are thinking of using a far more passive method of authentication. One method uses the actual WiFi signals in the air around you to identify you to your machines. This is done by measuring how the WiFi’s signal waves are behaving when they interact with objects such as the human body. There are a huge range of body shapes and motion patterns, even in related individuals, and these influence how WiFi signals behave and are received as they move around. Once the system recognises someone as being the person who lives in the home or is authorised to use the system, it can allow access to them for the entire internet-of-things.
The accuracy rate for this method was 75% with a single individual, and this increased to 95% for two people. The ideal number to maintain accuracy is 2-6 people in range of the router, which coincidentally is the size of most nuclear families. The researchers have named this system FreeSense. Now that the proof-of-principle has been demonstrated, the researchers want to improve the design. They want to increase the range, accuracy and number of people it can recognise.
This is all well and good, but both the researchers and the general tech community are missing out on a major fact. This may be a minor win for the home maker but it is also a major win for the home invader. Just a few tweaks and a bit of an impromptu software hack could see this technology turned into a way to track people inside their own homes. This would be especially useful for, say, a burglar. A tech-savvy thief could sit in a car in a street at night and use WiFi signals to try and find a house where no people are detected inside, or one where all detected people are upstairs in the bedroom (and therefore likely to be asleep).
The same goes for places of business – especially where valuables such as money or data are stored. One could learn patrol routes and real-time information of those guarding the premises all from the comfort of their car or home.
IoT lacks maturity
The technology is still in its infancy right now. Hopefully it will prove resilient to efforts from more untoward members of society who may wish to use it to profit at the expense of others and in opposition to the law. At the moment it may be a well-meaning attempt to increase security but in doing so may create more harm than good. It also displays how difficult it can be to balance security, ease of use and cost when trying to create secure systems suitable for public use.
Let’s remain optimistic
Misgivings aside, this technology also has a lot of silver linings and the possibilities could well be endless. A variety of new technologies which could spawn from it. Using WiFi to keep track of people could be useful in a community care sense where knowing if an elderly person has been lying on the floor for a while. It would also prove useful to detect if someone has not left the house for a number of days. These are just two examples of many.
As for potential security risks and FreeSense’s usefulness to criminals if adapted by them – as long as it can’t detect very large German Shepherds – then I think my property will be okay!