Recent years have seen the growth of the IoT. It is a concept that many now know, even if many do not fully understand its implications. Many see it as the ability of a smart thermostat to change the temperature by itself or a fridge to be able to re-order any food that may be out of date or finished. The truth is considerably more complex than that though, with the vast majority of the IoT’s power being unseen the vast majority. This could be sensors placed inside pipes, within engines, on lampposts, or anywhere else you can think of.
The challenge with the IoT has always been the placement of these sensors, they may have got smaller, but the reality is that getting them into the right places has always been difficult, especially when it comes to retrofitting previously dumb infrastructure to try and make it smart. After all, imagine trying to fit effective sensors to the inside of a victorian pipe in London that’s underneath a busy road and above a shallow tube line, not an easy task.
However, in the future this may not be an issue as smart dust may hold the key to being able to truly create the internet of things.
The idea of smart dust is not a new one, and the name was originally coined by Kris Pister in his 1997 paper ‘Smart Dust: Autonomous sensing and communication in a cubic millimeter’. The idea behind it is to create tiny sensors that can get anywhere, hence ‘dust’. This would mean that rather than having a sensor screwed to a lamppost and calibrated to a hub, you could theoretically paint several on when the lamppost has a coat of paint. Rather than needing to fit a sensor in an awkward Victorian pipe you could simply have them flowing through it with whatever the pipe carries.
Each can process and feedback data through a network to help monitor almost anything that traditional sensors can currently measure. At present the models being created by Pister have an accuracy rate of 99.99999%, which is impressive for what is essentially a computer that will could fit on the nail of your little finger. It gives the potential to increase the amount of data collected by orders of magnitude, with the potential for billions of sensors to be quickly and easily dispersed.
However, despite the concept and early experimentation being around since the mid 1990’s the technology is still some way off reaching its potential, but many companies are making furtive glances toward the emerging technology. 2016 was the first year that it appeared as the furthest left technology on the Gartner Hype Cycle, showing that it is making some progress in terms of widespread recognition. There is also some financial backing too, with Dust Networks, the company created by Pister in 2002, being acquired by Linear Technology in 2011 for $11 million, showing that there is clearly a belief that this will develop significantly in the coming years.
One of the biggest barriers has traditionally been powering the sensors, with early models hardly being dust sized, but 2 AA battery sized - because that’s what was needed to power them. They have gradually reduced in size as the sensors have become less power hungry and the means of powering them have become smaller. There is currently a lot of work taking place to allow them to become solar powered and they are also around the size of a grain of rice, considering smaller than the original 2 AA batteries.
The progress of smart dust is increasing and with public awareness of it increasing this progress is only going to accelerate, and who knows where it could end up.