IEEE Consumer Electronics Magazine - January/February 2024 - 55

& ADDITIVE MANUFACTURING (AM) is the process
of developing 3D objects through an additive,
layer-by-layer approach.1 This process, also
known as 3D printing, provides fast prototyping
of complex parts and can utilize food, polymers,
metals, and living tissue or matter.2 As
the complexity and efficiency of 3D printing
improves, additional features (including thirdparty
specialized parts) are required for
higher integration and more efficient automation.
These additional components and communication
methods require a robust and strong
communication path that minimizes cost of AM
and 3D-printing devices. Controller area network
(CAN) has recently been increasing
in popularity in 3D printers and other embedded
system-based AM machines. Originally
designed for vehicular electronic systems, the
CAN bus has also found use in other industries,
such as aerospace3 and robotics.4 Unlike traditional
communication protocols, CAN provides
fast, low-error transmission on a two-wire differential
bus shared among all devices, allowing
easy modularity, providing strong EMI protection,
and significantly reducing wire cost. Thus,
many printers have begun utilizing this alternative
to the traditional protocols to optimize the
cost and expand-ability of their application.
However, CAN's shared, unencrypted communication
bus has previously led to many proposed
attacks targeting vehicular electronic
systems. While vehicles are the primary target of
CAN-related attacks, these huge vulnerabilities
are also extremely exploitable within CAN AM
systems. CAN's destination-based IDs mean that
any controller can use another connection's ID
pattern to easily transmit and receive untraceable
messages without any authenticity needed.
This means that one compromised node on
the CAN network in a 3D printer or printer
farm could compromise the entire system's
communication.
Thus, discussion of these attacks in context
to 3D printing is important to fully understand
and counter vulnerabilities as CAN usage continues
to grow in AM. An understanding of the capabilities
of hardware-related attacks will further
promote the use of CAN security frameworks in
professional environments, including commercial
printers and AM farms. Unlike vehicular CAN
January/February 2024
FIGURE 1. Concept of CAN bus hacking through a
compromised node, either via hacking an existing
node or installing a hardware trojan.
systems, the modules being connected to a 3D
printer change regularly, thus greatly increasing
the exposure and surface of attack from
adversaries.
Our contribution in this work is to discuss
and demonstrate the capabilities of CAN attacks
on AM systems that utilize CAN networks for
communication between parts/modules. We also
evaluate the vulnerabilities and attack surface of
a commercial 3D printer that uses a CAN network.
In addition, this article will describe a generalized
attack structure that an attacker/
eavesdropper would use to perform these
attacks discretely through means of compromising
a node or new component with malicious
hardware, as shown in Figure 1. By utilizing a
module's physical connection to the bus, attackers
can control and manipulate information from
all other modules. Only one module vulnerability
must be exploited to take control of all bus communication.
With this bus-based connection, any
information can be sniffed or spoofed from servos,
motors, printer head sensors (such as thermistors),
controllers, limit switches, and more.
Thus, by compromising a module or installing
seemingly benign modules containing malicious
hardware in unprotected/insecure CAN systems,
the many attacks described in this work can be
performed against CAN-based 3D printers. The
rest of this article is organized as follows. The
" Background and Related Attacks " section provides
a background on CAN attacks. The " Attack
Structure and Demonstration " section describes
the attack structure and procedure to identify a
point to attack, and also provides a
55
IEEE Consumer Electronics Magazine - January/February 2024

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - January/February 2024
