Ordinary Smartphones Able to Hack 3-D Printers

(kaboompics.com/Pexels)

8 September 2016. A hacker with an everyday smartphone can steal design data from three-dimensional printers, according to a new paper from a university computer science group. A team from the Embedded Sensing and Computing Lab at University at Buffalo in New York describes its discovery next month at the ACM Conference on Computer and Communications Security in Vienna, Austria.

“Many companies are betting on 3-D printing to revolutionize their businesses, but there are still security unknowns associated with these machines that leave intellectual property vulnerable,” says lab director and computer science professor Wenyao Xu in a university statement. In their paper, Xu and colleagues show how built-in features on today’s mobile phones can detect and read signals given off by 3-D printers.

Many of today’s s printers in additive manufacturing — industrial applications of 3-D printing — have security features such as encryption, and can produce watermarks to foil pirating. The Buffalo team exploited a different vulnerability, constructing a side-channel attack, a form of hack that reads emissions from electronic devices, then reconstructs the signals to reveal the nature of the work being performed.

In this case, the authors were able to program built-in sensors on a smartphone, a Nexus Android model, to detect and capture signals within 20 centimeters (about 8 inches) from an Ultimaker 2 Go device, a popular open-source 3-D printer. The phone detected and measured changes in sounds made by the printer nozzle as it started and stopped, extruded material, and moved vertically and laterally. In addition, the phone’s magnetometer sensor detected and captured changes in magnetic fields emitted by the printer. The researchers developed algorithms to integrate these two data streams, then statistically process the data to reconstruct the G-codes, control language instructions directing actions of the printer.

Tests of their side-channel attacks show the team could reconstruct simple rectangular or near-rectangular objects within about 6 percent of the original design. Reconstructions of complex objects, resembling items such as medical devices or auto parts more likely to be produced in additive manufacturing plants, come within about 10 percent of the originals.

Xu and colleagues offer suggestions for improving security of 3-D printers to counter side-channel attacks. Keeping mobile devices further from the printer reduces the detection of signals; at 40 centimeters for example, error rates increased to about one-third. Physical barriers can shield electromagnetic emissions and ambient nose can disrupt the capture of acoustic signals. Increasing the print speed also makes it more difficult to determine the movement of the printer’s nozzle.

Read more:

• Computer Models to Boost 3-D Print Designs in Mfring
• Univ. Spin-Off Develops New Cybersecurity Paradigm
• Univ. of Pittsburgh, Ansys Partner on Additive Mfring
• Wearable Personal Instruction System Being Developed
• 3-D Printed Guide Devised to Regrow Nerve Fibers

*     *     *