Researchers from the Massachusetts Institute of Technology (MIT) have developed processors capable of providing encryption to low-power devices.
The Internet of Things (IoT), including embedded industrial sensors, smart lighting, and monitoring systems, can be invaluable for the enterprise.
These devices tend to have minimal power demands, can operate on Low-Power Wide-Area Networks (LPWAN), and have entered the enterprise, industrial scene, and consumer realm rapidly over the past few years.
However, when there is little power to draw upon, security can sometimes be an issue as there are no resources or means to implement traditional security solutions.
Encryption, for example, is a sticking point for today’s IoT devices. While many of our online transactions and communication are protected by public-key cryptography, many IoT devices cannot support individual encryption implementations due to a lack of power and memory.
However, the encryption challenge may have been solved by MIT researchers.
A team of electrical engineers and computer scientists have developed a new chip, hardwired for public-key encryption, which consumers only 1/400 of the power standard protocols usually consume.
In addition, the processors only use 10 percent as much memory as once required for encryption, and the researchers say can also execute commands up to 500 times faster.
The chip uses what is called elliptic-curve encryption. This relies on a range of mathematical functions and curves to enhance security — and MIT’s chips have been designed to handle any curvature.
The team has hard-wired the datagram transport layer security protocol, which handles the formatting and transmission of encryption, into the chip design alongside a processor which is able to handle additional elliptic-curve-based security protocols. However, this processor can be powered down when not in use to increase energy efficiency.
A custom inverter circuit also increases usable surface area by 10 percent while reducing energy consumption by 50 percent.
“Cryptographers are coming up with curves with different properties, and they use different primes,” said Utsav Banerjee, an MIT graduate student and author on a paper documenting the research. “There is a lot of debate regarding which curve is secure and which curve to use, and there are multiple governments with different standards coming up that talk about different curves.”
“With this chip, we can support all of them, and hopefully, when new curves come along in the future, we can support them as well,” Banerjee added.
The findings will be presented in a paper this week at the International Solid-State Circuits Conference.
In September, IBM and MIT joined forces to establish a new research facility in Cambridge, the MIT-IBM Watson AI Lab, with a focus on artificial intelligence, healthcare, and cybersecurity.