Embedded device security has come a long way since the days of telnet and default passwords. Product vendors are now doing more to secure their devices but how effective are they? This presentation will outline many of the software and hardware-based attacks used to compromise embedded systems. It also discusses some of the mitigations used to prevent these attacks. Many previous IoT talks show the simplicity of hacking devices that have weak security or no hardening. In contrast, this presentation shows how even secured devices have attack surfaces that still need to be addressed. It demonstrates the need for embedded devices to incorporate a security lifecycle plan and hardware designs must be audited for security weakness before production. Topics to be covered include firmware image encryption, disabling UART console access, hardening JTAG development access, securing e.MMC storage, NOR Flash protection, processor glitching, update lifecycle attacks, avoiding custom crypto, dealing with reverse engineers, and initial device setup vs authentication. None of these topics will be a deep dive. The intent is to show how they are attacked or utilized to mitigate specific attacks. To illustrate these topics the presentation will use a recent security audit of a US solar equipment manufacturer as a case study. The vendor incorporated many best practices for securing embedded devices but made some architecture decisions in the guise of security that ended up weakening their security posture rather than helping it. Finally, we'll show the ramifications of an attack against solar systems and how it could be used for racketeering. Attacks in this talk are beneficial to system designers, hobbyists, and researchers.
Get notified about new features and conference additions.