In the beginning of the modern computer’s history, it used to be true that these things were behemoths. A single server took up an entire room on its own, with barely enough storage for a simple JPEG picture. The equipment in total weighed several tons and getting anything out of it used to take days. Today, that’s no longer the case, with your average smartphone far outrunning the capabilities of those early dinosaurs. The trend with computers now is to make them smaller and more efficient, and engineers at the University of Michigan School of Engineering reckon they have the world’s smallest computer record tied down for a while. Meet the Michigan Micro Mote (M³)!
Most variations of the Micro Mote measure one millimeter cubed. It is ridiculously small, with the team working on the design spending just over ten years perfecting it. The project is headed up by Computer Science and Engineering Professor David Blaauw, which was initially started by one of Blaauw’s students who just wanted to see how small of a battery the team could make, and how low they could scale the voltage and still have enough power to have a working nanoscale computer.
While commonly found low-voltage computers in your smartphone or pace maker operate on the milliwatt scale in terms of energy usage, the Micro Mote operates on the nanowatt scale. If you had to represent how much of a change this is, it would be like comparing the surface area of a sheet of A4 paper to a standard football field.
Over time, the goal of the M³ project extended into developing “a complete wireless sensing and computing node that includes a low-resolution imager, signal processing and memory, temperature sensor, on-board CMOS timer, wireless communication, battery, and solar energy harvesting that are all packaged in a 1mm3 volume through low-cost die stacking and encapsulation.”
Science fiction has been forever making us dream of miniature computers that are small enough to go just about everywhere. There are nanomachines in Metal Gear Solid that do all sorts of things. In the latest episode of the CW TV series remake of “The Flash“, Ray Palmer has a programmable microbot injected into his blood stream to remove a blood clot that is on the way to give him brain damage. Something like the Micro Mote comes close, but its still not small enough to enter our blood stream just yet. However, it is small enough to be embedded under the skin, offering a wide range of capabilities like NFC at your fingertips, or relaying information to a medical station that is monitoring your health.
The most interesting part of the Micro Mote is the battery. Sized down to near-microscopic levels, it holds its charge for several weeks, powering the low-energy RISC processor that powers the computer and runs any of the optional sensors. Charging the battery is done by holding the computer under a strobing light running at a high frequency, which also facilitates the sending and receiving of data for programming the computer. When not embedded deeply into something, the battery can be charged using solar power.
When its not being charged from the strobing light, the Micro Mote communicates with a remote station using radio frequencies, reducing the effective distance to about five meters. In the future, Blaauw and his team imagine that the Micro Mote will be used in an array of industries, including low-power, massively parallel servers that carry hundreds of thousands of these things in a single 1U rackmount chassis.