Forensics of Thermal Side-Channel in Additive Manufacturing Systems

Additive manufacturing systems, such as 3D-printers, emit cyber-data via physical side-channels (such as acoustic, power, thermal, and electromagnetic emissions) while creating 3D objects. These emitted data can be used by attackers to their advantage for indirectly reconstructing the 3D objects being printed along with its corresponding cyber-data. Moreover, in our work, we demonstrate that the thermal emission can be taken as one of the side-channels to monitor the leakage of the cyber-data from the 3D-printer. This is an example of a physical-to-cyber domain attack, where information gathered from the physical-domain can be used to reveal information about the cyber domain. Our novel attack model consists of a pipeline of image processing, signal processing, machine learning algorithms, and context-based post-processing to improve the accuracy of the object reconstruction. In our experiments, we have gathered data using a thermal camera and partially reconstructed the object as a proof of concept of our attack model. Our work exposes a serious vulnerability in additive manufacturing systems exploitable by physical-to-cyber attacks that may lead to theft of Intellectual Property (IP) and trade secrets. To the best of our knowledge this kind of attack has not yet been explored in additive manufacturing systems.