Skip to Content

A Chip for Longer-Lasting Wearable Computers

Batteries for smart watches and other wearables never last long. A new microchip design could change that.
August 27, 2014

Existing wearable devices such as Google’s head-mounted computer Google Glass require battery charging at least once a day, even with light use. A new kind of low-power chip aimed at such wearable devices could not only extend battery lives but also allow the devices to constantly listen for voice commands.

The new chip, made by startup company Ineda Systems, is intended to work alongside the main processor inside a device, performing functions such as listening for voice commands and running simple apps. That saves energy by allowing the main processor to spend more time powered down.

“We looked at the typical use cases of a wearable device and we designed a chip with those in mind,” says Ajith Dasari, vice president for platforms and customer engineering at Ineda. “About 90 percent of the time a user’s device will be in ambient mode or they will be only using simple apps.”

Ineda is currently testing samples of two chip designs and says it aims to move them into mass production sometime next year. The company was founded in 2010 and is backed by investors including Qualcomm, which dominates the market for chips inside smartphones and tablets, and Samsung.

Ineda’s chips feature either two or three processor cores. One core has relatively little computing power but also consumes very little power and is designed to be always operating. The other one or two cores on an Ineda chip have more power and are only switched on for heavier tasks. If that’s not enough for the job, the chip wakes up the device’s main, power-hungry processor as a last resort.

Ineda’s most complex chip design, known as Advanced, is aimed at high-end smart watches and should enter production next year. The lowliest of its three cores can perform tasks such as monitoring motion sensors to look for gestures, maintaining a Bluetooth connection to another device, and recognizing a key spoken phrase that wakes up a device. The chip’s second core powers up alongside the first for more complex tasks such as playing music, recognizing a handful of voice commands, or running a simple app such as a heart-rate tracker. Once the third core joins in, the chip can perform more complex tasks, such as full speech recognition, that require accessing data over the Internet.

A less powerful design with only two cores, known as the Micro, is also in testing and headed to production. It is intended for less feature-rich watches and wearable devices, and could also be added to a smartphone to make it more power-efficient, says Dasari.

Smartphones may be where Ineda’s chips first appear, because some manufacturers have already experimented with adding helper chips to improve battery life and functionality. The latest iPhone, for example, includes a chip dedicated to processing motion sensor data (see “What Apple’s M7 Motion-Sensing Chip Could Do.” Motorola’s Moto X phone has a similar chip, as well as another that constantly listens for the phrase “OK, Google” (see “The Era of Ubiquitous Listening Dawns”). Ineda’s chips can take on a wider range of tasks, says Dasari. “The Moto X has two different chips, but we can achieve the same with one chip,” he says.

Tulika Mitra, an associate professor at the National University of Singapore, says that Ineda’s approach will help save power. ARM, the company whose designs underpin most mobile device processors, has moved in a similar direction by releasing a line of “big.LITTLE” chips with large and small processor cores, she says. But Ineda has taken that line of thinking further. “Instead of only big and little cores, now you have a range of cores.”

One challenge for Ineda’s chips is that operating systems used by mobile device manufacturers today would likely need some modification to support its novel multi-core architecture, says Mitra.

Dasari acknowledges that software integration is an issue for Ineda, but says the company is working on tools that could address it.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.