The CAN data link layer protocols (CC, FD, and XL) as standardized in ISO 11898-1:2024 do not provide security measures. If needed, they must be added. According to the OSI (open systems interconnection) model, security controls can be applied to each of the seven layers, depending on the required security level and expected attack scenarios. This is standardized in ISO 7498-2:1989 (Information processing systems — Open Systems Interconnection — Basic Reference Model — Part 2: Security Architecture).

CiA has a long history in developing security measures regarding unintended misuse as well as intended manipulation of CANopen communication. This covers also unauthorized access to CANopen devices and CANopen networks. There are password options for the CANopen object dictionary access. Furthermore, there is an authentication signature option in CANopen messages specified, indicating that they are from the right origin and that in case of SDO (service data object) transmission, segments belonging together and that they have not been manipulated. Some CANopen profile specifications provide even dedicated security measures including parameter encryption.

In ISO 14229-3, there is standardized a cybersecure CAN-based interface using unified diagnostic services (UDS). This standard is used for road-vehicle diagnostics, for example. Some automakers apply this standard also for software downloads and updates.

There are also proprietary security controls for CAN-based communication on the market. This includes CAN transceivers monitoring that the assigned CAN IDs (identifier) are not used by CAN frames transmitted by other nodes. In this case, the CAN transceiver destroys those data frames by means of sending CAN error frames.

Supporting cybersecurity controls in embedded systems typically involves a trade-off between the required processing resources and security measures needed for attack vectors in a dedicated application. In order to assist the CAN community and to meet latest cybersecurity requirements for embedded systems, CiA has started to develop several cybersecurity measure building blocks.

One is a data link layer add-on protocol to be implemented in the CAN XL protocol controller hardware (also known as CANsec®, which is a European trademark by CiA). Other building blocks are higher-layer protocol (HLP) approaches suitable for CAN FD based networks. There is also a new HLP solution suitable for CANopen CC (classic), in preparation. It is also intended to specify a secure configuration method for CANopen devices.

Additionally, CiA has established the IG08 Ethernetification of CAN, which specifies the mapping of secured and compressed Ethernet frames (MACsec protected frames) to CAN frames. In the first step, a mapping to CAN FD (flexible data-rate) data frames is planned. This approach is intended for automotive applications, but not limited to them.

Furthermore, some CiA members propose the development of a higher-layer management protocol to be embedded into CAN FD data frames. This includes the indication of an optional higher-layer security measure. This means, the CAN FD data frame payload (data field and/or identifier field) contains information, whether the remaining payload is protected against attack.

At the end of 2025, CiA plans to publish a guideline for designing CAN-based secure network systems and cybersecure devices respectively ECUs (electronic control units) with CAN connectivity. CiA also observes CAN-related cybersecurity specifications and recommendations by international standardization bodies (i.e., IEC and ISO) and other nonprofit consortia (i.e., Autosar, Open Alliance, and SAE), in order to reference them, where appropriate.