FPGA Implementation of HLS crypto accelerators for embedded security in autonomous vehicles

The growing ubiquity of autonomous vehicles (AVs) has introduced a new attack surface for malicious actors: the embedded systems that govern a vehicle's critical operations. Security breaches in these systems could have catastrophic consequences, potentially leading to loss of control, manipulation of sensor data, or even physical harm. To mitigate these risks, robust cybersecurity measures are paramount. This research delves into a specific threat – side-channel attacks – where attackers exploit data leakage through unintentional physical emanations, like power consumption or electromagnetic waves, to steal cryptographic keys or sensitive information. While various software and hardware countermeasures have been proposed, this study focuses on the implementation of masking techniques within the realm of embedded security. Masking techniques aim to obfuscate sensitive data during cryptographic operations, making it significantly harder for attackers to exploit side-channel vulnerabilities. This research explores the suitability of a Boolean masking approach within a high-level synthesis environment. This system-level approach offers several advantages over traditional design methodologies. It facilitates faster design processes by enabling early identification and rectification of errors. Additionally, it streamlines hardware-software co-design, allowing for a more integrated and efficient security architecture within the autonomous vehicle's embedded systems. Furthermore, the system-level approach enables the application of advanced validation strategies, ensuring the effectiveness of the implemented masking techniques. To evaluate the efficacy of the Boolean masking approach, the research investigates its application to three prominent block cipher algorithms – PRESENT, AES, and Serpent – all of which are based on substitution-permutation networks (SPNs). By implementing these masked algorithms in C and simulating their performance within an embedded system context, the study assesses factors like resource utilization and overall processing speed. This comparative analysis aims to identify the most effective masking implementation for protecting cryptographic operations in autonomous vehicles. Ultimately, the findings of this research can inform the development of robust security frameworks that safeguard autonomous vehicles against side-channel attacks and other cyber threats, paving the way for a safer and more secure future of transportation.