The tiny (yet powerful) world of speckled computing
Networks of small devices that can sense, compute and communicate data emerge from lab
We are inching closer to a world in which everything -– and everybody -– will be part of some wireless computer network.
Or, more precisely, a Specknet. "Specks" are tiny computing devices that can be placed on common objects -- or on people, as they currently are in a Scottish community testing “speckled computing” with respiratory patients.
Each individual programmable speck can sense real-world data (temperature and motion, for instance), compute it, and communicate with each other wirelessly.
The goal, as laid out by the Centre for Speckled Computing at the University of Edinburgh in Scotland, is no less than to bridge the physical and virtual worlds:
“In our vision of Speckled Computing, the sensing and processing of information will be highly diffused – the person, the artefacts and the surrounding space, become, at the same time, computational resources and interfaces to those resources."
In the case of the Scottish trial patients, small electronic patches (about the size of a man's thumb) placed on their chests monitor respiratory data -- a person’s breathing rate and depth, for example -- and transmit the information wirelessly to doctors monitoring the patients miles away.
This community-care pilot project followed successful hospital trials of the new technology in Scotland based on a system created at the Centre for Speckled Computing, headed by Professor D K Arvind. (See interview below.)
Speaking this month at the annual British Science Festival in Aberdeen, Scotland, Dr. Arvind explained that the “inspiration” for the speckled computing model comes not from the virtual or digital world, but from the living world.
“A large number of these is almost like a semblance of cells and neurons,” he said. “It’s very similar to how we believe biological computation takes place.”
The vision of a world in which computers are embedded in everyday objects has been around a long time. Way back in 1988, Mark Weiser, the late, renowned Xerox PARC scientist, was talking about this very notion, which he called “ubiquitous computing” and which he predicted would become as common and invisible as electricity.
“Hundreds of computers in a room could seem intimidating at first,” Weiser wrote in 1991. “But like wires in the walls, these hundreds of computers will come to be invisible to common awareness. People will simply use them unconsciously to accomplish everyday tasks.”
Speckled computing essentially is ubiquitous computing in granular, dispersed form. But it wouldn’t be possible without advances in Internet technology. As Arvind explained in his inaugural lecture for the Centre for Speckled Computing earlier this year, Internet Protocol 6, only now just being rolled out, eventually will support 35 trillion subnetworks, each able to connect millions of devices.
Much of the early focus on uses for speckled computing has targeted health care. Besides using specks to allow doctors to monitor respiratory patients, researchers are investigating using them to help prevent falls among the elderly and to measure conversational skills in adults with Asperger Syndrome.
Beyond healthcare, speckled computing appears to have numerous fascinating potential applications. Among them are:
* Enabling humans to control the actions of robots with their own body motions, so a human raising his speck-covered arm would be able to make a robot tied into the same Specknet raise its arm in an identical manner. This would have potential uses by the military and rescue personnel.
* Increasing the ability of scientists to understand “land-atmosphere interactions” such as fires. Specknets may be used to monitor seasonal cycles and their impact on land surfaces, and can also be connected to simulation models.
* Allowing people “hands-free” and “eyes-free” control of devices via body movements, including head, wrist and foot gestures.
* Measuring stress and forecasting disease for agricultural crops
If you have time and are interested in learning more about speckled computing, Dr. Arvind's inaugural lecture on YouTube is a good place to start. However, if you don't have 70 minutes to spare at the moment, you can learn a lot about speckled computing directly from Dr. Arvind, who was kind enough to answers some of my questions via email:
From the Lab: Specks currently are small, but still easily visible. How much smaller can they get? As small as grains of rice or sand?
Dr. Arvind: The limiting factor in the miniaturisation of specks is not the electronics but the battery sizes required to do anything practical. Improvement in energy density of batteries is quite modest compared to the doubling of transistors every 18 months. Also, as we make the specks smaller the radio frequency for wireless communication has to move up the GHz range (24GHz compared to the current 2.4GHz) so that the antenna size gets smaller at the cost of increased power consumption. An alternative is optical communication using laser/LEDs which is low power but with the disadvantage of fixed specks for line of sight communication and problems of occlusion.
From the Lab: What are the most promising areas for speckled computing?
Dr. Arvind: We are concentrating on three application areas for specks: Healthcare, environmental monitoring and digital media.
In healthcare, we have developed the Respire speck which can monitor respiratory rate/flow, heart rate, and activity levels. It has undergone clinical trials in Edinburgh hospitals for postoperative care in hospitals and has been deployed in the community to monitor remotely COPD patients in the comfort of their homes. The Orient speck has been used in mobile gait analysis and remote physiotherapy.
On environmental monitoring, we are developing specks with GPS, accelerometer, and magnetometer to analyse the behaviour of wild horses in southwest Spain. The information of interest are the areas visited by the horses, times spent resting, running, feeding and their social behaviour -- which horses spend time with others which is useful to study the spread of disease. The specks are currently undergoing trials in a herd of horses (not wild) in Edinburgh and will be deployed in Spain in 2013.
Other applications are the monitoring of the environment in buildings and greenhouses to optimise usage of energy.
In digital media, we have developed on-body wireless specks for full-body, 3-D motion capture in real-time. This has applications in 3-D animation, user interfaces, biomechanics, dance, sports (analysis of golf swing).
From the Lab: What are the greatest practical challenges to speckled computing? Cost? Power supply? Wireless reliability?
Dr. Arvind: The greatest challenge is taking great care in the design of the speck architecture, firmware, networking protocols, distributed algorithms so as to optimise energy usage.
From the Lab: Is there any danger of a Specknet being compromised? Are they inherently insecure?
Dr. Arvind: Not necessarily. The data transmitted by specks can be encrypted for wireless transmission which is an overhead, of course.
From the Lab: How long do you think it will be before “ubiquitous computing” truly is as common as electricity?
Dr. Arvind: We have come a long way in the last 10 years. Speckled computing presaged the "Internet of things" by a good many years! When the research community was talking in terms of wireless sensor networks, it was clear to me that it was not the sensor data but the analysis of the data on the specks, at the edges, to extract information in situ was the key and its connection to the rest of the IP network with the advent of the IPv6 protocol and the accompanying explosion in addresses. I do believe we will get there but cannot be precise when.