Smartphone can hack 3D printers

RESEARCHERS have used sensing equipment on smartphones to measure “leaked” energy and acoustic waves generated by 3D printers, demonstrating the potential to steal intellectual property.

A team from the University at Buffalo (UB), US warned of the vulnerabilities of 3D printing – or additive manufacturing – in a study that explores the security of a potentially multibillion dollar industry that can be used to create a range of objects including rocket engines and heart valves.

Previous studies, including a warning from New York’s Tandon School of Engineering in July, have simulated a cyberattack using the internet to hack into a 3D printer from the outside to make changes to the CAD design files.

In this study, the team did not simulate a cyberattack, instead they programmed a common smartphone’s built-in sensors to measure the electromagnetic energy and acoustic waves emanating from a 3D printer in operation. These sensors can infer the location of the print nozzle as it moves to create the object being printed.

At a distance of 20 cm from the printer, the phone gathered enough data to enable the team to replicate the printing of a simple object, such as a door stop, with 94% accuracy. For complex objects, such as an automotive part or medical device, the accuracy was lower, but still above 90%.

Kui Ren, a professor in UB's Department of Computer Science and Engineering, said: “The tests show that smartphones are quite capable of retrieving enough data to put sensitive information at risk.”

80% of the useful data come from measuring the electromagnetic waves and the remaining data came from acoustic waves.

With the commonality of smartphones, this study could be very useful to unsuspecting engineering industries that are moving to rely on 3D printing for manufacturing using protected intellectual property. The team say they have safeguarding measures to prevent disgruntled employees or industrial spies from gaining access to the data.

The team say the simplest solution to prevent theft is distance, as the ability to obtain accurate data for simple objects diminished to 87% at 30 cm, and 66% at 40 cm.

Another solution is to increase the print speed. The team said that emerging materials may allow 3D printers to work faster, making it more difficult for smartphone sensors to determine the print nozzle's movement. The team also recommend software-based solutions, such as programming the printer to operate at different speeds, and hardware-based ideas, such as acoustic and electromagnetic shields for companies with particularly sensitive data.

The team will present the research at the Association for Computing Machinery's 23rd annual conference on computer and communications security, next month in Austria.

A copy of the study can be obtained by emailing Cory Nealon, UB university communications, cmnealon@buffalo.edu.