Why Embedded Security Fails, and What to Do About It
(Room 303C)
05 Nov 25
3:25 PM
-
3:50 PM
Tracks:
Embedded Security - Attacks & Hacks
Secure cryptographic algorithms. License checks. Encrypted firmware updates. Enabled firmware readout protections. On paper, many embedded systems tick the right boxes for security. But without adversarial thinking, these protections often fail under real-world scrutiny - attackers can reverse engineer the binary and defeat them with minimal effort.
This talk explores how embedded security goes wrong – even when using the “right” technologies. Through practical, real-world inspired examples, we follow a typical attack path: from firmware extraction to reverse engineering and exploitation.
We’ll demonstrate how attackers extract hardcoded secrets like AES keys or RSA keys using only a disassembler and a string search. Then we’ll examine how license checks, and feature restrictions are easily bypassed through binary patching and how proprietary algorithms are reverse engineered using open source tools like Ghidra. Finally, we’ll explain why coding in C/C++ without proper exploit mitigations—such as control-flow integrity or stack canaries—leaves devices open to memory corruption attacks, and why enabling these mitigations on embedded targets is often more difficult than it seems.
We conclude by outlining practical, binary-level protections - such as obfuscation, integrity checks, anti-debugging, and exploit mitigations – that raise the bar for attackers, without requiring source code or hardware changes.