Please assign a menu to the primary menu location under menu

Mobile news

Future Phones Could Be as Easy to Build and Upgrade as LEGOs


Courtesy the researchers and edited by MIT News

Courtesy the researchers and edited by MIT News

If you’ve ever wanted to upgrade the hardware for your phone or smartwatch without having to throw it out with every new product rollout (I’m looking at you, Apple), the future is right around the corner.

Engineers at MIT have made a wireless, reconfigurable chip that one day could easily be snapped onto your existing devices like a LEGO brick. In a paper published Monday in Nature Electronics, the chip contains layers of sensing and processing elements—including its own neural network (a computer that simulates the brain)—that can be programmed to do whatever you want, whether it’s tracking muscle activity or adding an image recognition feature to your device. Having the flexibility to customize and upgrade an old device is a modder’s dream, but the chip may also help reduce electronic waste, which is estimated at 50 million tons a year worldwide.

“We tried to make some kind of AI chip, which is reconfigurable, so we call it a LEGO-like AI chip,” Jihoon Kang, an electrical engineer at MIT and co-author of the new study, told The Daily Beast.

Future Robots Could Wear This Living Skin That’s Self-Healing

Kang explained that in most electronic devices, chips are hardwired to suit the needs of that particular device so there’s no way of reconfiguring if you wanted to improve or upgrade.

To get around being set in electronic stone, Kang and his colleagues came up with a tiny chip that transfers data wirelessly through light. Sensors, photodetectors, and LEDs sitting on top of the chip send information through light signals to the chip above it.

The chip does have an artificial brain—a computing core no bigger than four square millimeters composed of what engineers call memristors, which remember how much electric current last flowed across. These memristors function together to form a kind of “brain-on-a-chip,” interpreting those light signals and executing some pre-programmed function, said Kang.

Right now, the chip can carry out some basic image recognition tasks, recognizing pixelated forms of the letters M, I, and T. It had some slight difficulty differentiating the M from I and T—Kang says it has to do with the image getting slightly corrupted when it’s picked up by the sensor—so the researchers are working on improving the reliability of the data transfer.

Watch This Bionic Butterfly Change Colors as Its Wings Flap

But in the future, Kang and his co-author, Min-Kyu Song, also an electrical engineer at MIT, are hoping they could integrate their chip with a “smart skin” their team previously developed to monitor vital signs.

“Electronic skin can measure different types of electrophysiological signals such as [brain waves], heart signals, muscles signals, and others,” Song told The Daily Beast. “One of our projects is to attach this ‘smart skin’ to the forehead and measure brain waves to estimate the stress levels of the user and integrate with the AI chip.”

Kang and Song said it may be possible in the future to use their AI chip for other wearables, smartwatches, and even augmented reality glasses. But they believe the most benefit would be to the average consumer where someone could essentially buy a library of LEGO-like AI chips with various pre-programmed functions that could be stacked on top of each other and work in synergy.

“We can just sell different types of AI networks, like a videotape or CD, then you can freely choose which AI you want,” said Kang. “Just replace the chip for all their AI needs.”

Read more at The Daily Beast.

Got a tip? Send it to The Daily Beast here

Get the Daily Beast’s biggest scoops and scandals delivered right to your inbox. Sign up now.

Stay informed and gain unlimited access to the Daily Beast’s unmatched reporting. Subscribe now.



READ SOURCE

Leave a Reply

This website uses cookies. By continuing to use this site, you accept our use of cookies.