Could RFID Be the Solution to Battery-less IoT? Arm Thinks So
Coming up on its 30th anniversary, Arm is now working on ways to create a battery-less future for the IoT.
In almost every discussion about the IoT, it's glaringly obvious that the major challenge facing the future is how to power these devices. Briefly, IoT products are mostly designed to be battery-powered standalone devices, deployed somewhere in which changing the battery is mostly unfeasible. At the same time, these devices are expected to perform significant computations, sometimes even hosting ML applications on board.
The challenge to extend battery life shapes up in two ways: designers must either develop more energy-efficient hardware or develop sustainable ways to power the devices. For the latter, Arm is optimistic about a new application of an old technology: RFID.
Passive RFID tags are a popular form of RFID that consumers interact with daily, used in applications like credit cards. The exciting thing about passive RFID is that it doesn’t contain a battery: instead, it harvests electromagnetic energy from the device used to read it—that’s why we don’t have to worry about recharging our credit cards.
RFID in credit cards. Image used courtesy of Arm
This is good and well, but passive RFID tags are limited in what they can do. While these devices are capable of receiving and storing information, that’s the end of their capabilities. RFID can’t do anything else with the data besides sending and receiving.
For RFID to be used in IoT, someone will need to marry the low-power, battery-less potential of RFID with the computational needs of IoT.
A newly-announced initiative by Arm, dubbed Project Triffid, hopes to exploit characteristics of RFID tags for use in IoT applications.
The solution lies in developing what Arm calls “computational RFID” or CRFID for short. The idea is that, theoretically, it’s possible to embed a very low-power or even battery-less MCU suited for IoT applications inside of an RFID tag. This way, the tag would be able to both capture and store information, like conventional passive RFID, while also acting upon it.
Diagram of the difference between RFID and computational RFID. Image used courtesy of the Benjamin Ransford et. al
"CRFID systems use scavenged radiofrequency energy emitted by ceiling- or drone-mounted readers (for example), and can be suitable for deployment in environments where other harvesting systems may fail," explains Arm engineer James Myers.
Developing this technology in practice will be complicated. In a far off deployment, where will the tags harvest energy from? Engineers at Arm say that their desired design specs are a 10-meter range from the reader, giving the device a 1uW power budget.
What Will It Take To Get There?
According to engineers at Arm, the company has previously built battery-based systems that can operate at these low power levels. However, the task becomes significantly trickier when employing energy harvesting, introducing regular and random power failures that may happen hundreds of times per second.
Types of energy harvesting technologies. Image used courtesy of Xiaoli Tang et. al
Project Triffid will need to develop circuits to detect these irregularities, software that can react and adapt to complications, and non-volatile memory to store the work.
The company is in the process of testing 28nm chips and plans "to tape out the next iteration of chips at 22nm, enabling the incorporation of MRAM."
A Battery-less Future?
In a world where powering IoT devices is still a big question mark, this solution by Arm holds significant implications. Arm believes that the potential use cases for Project Triffid are extensive, from tracking and testing the safety of perishable goods to monitoring product lifecycle (for instance, in shoes) to ensure the product returns to a manufacturer for recycling.