Saurav Maji

Saurav Maji, Utsav Banerjee, Anantha P. Chandrakasan

With the recent advancements in machine learning theory, many commercial embedded micro-processors use neural network models for a variety of signal processing applications. However, their associated side-channel security vulnerabilities pose a major concern. There have been several proof-of-concept attacks demonstrating the extraction of their model parameters and input data. But, many of these attacks involve specific assumptions, have limited applicability, or pose huge overheads to the attacker. In this work, we study the side-channel vulnerabilities of embedded neural network implementations by recovering their parameters using timing-based information leakage and simple power analysis side-channel attacks. We demonstrate our attacks on popular micro-controller platforms over networks of different precisions such as floating point, fixed point, binary networks. We are able to successfully recover not only the model parameters but also the inputs for the above networks. Countermeasures against timing-based attacks are implemented and their overheads are analyzed.

Saurav Maji, Utsav Banerjee, Samuel H Fuller et al.

This paper presents a dual-factor authentication protocol and its low-power implementation for security of implantable medical devices (IMDs). The protocol incorporates traditional cryptographic first-factor authentication using Datagram Transport Layer Security - Pre-Shared Key (DTLS-PSK) followed by the user's touch-based voluntary second-factor authentication for enhanced security. With a low-power compact always-on wake-up timer and touch-based wake-up circuitry, our test chip consumes only 735 pW idle state power at 20.15 Hz and 2.5 V. The hardware accelerated dual-factor authentication unit consumes 8 $μ$W at 660 kHz and 0.87 V. Our test chip was coupled with commercial Bluetooth Low Energy (BLE) transceiver, DC-DC converter, touch sensor and coin cell battery to demonstrate standalone implantable operation and also tested using in-vitro measurement setup.