Davide Giacomo Marangon

Over the last decades, Quantum Key Distribution (QKD) has risen as a promising solution for secure communications, a pressing subject in the aftermath of the security threat posed by Quantum Computers and the Shor's Algorithm. Offering a theoretically secure way to share secret keys between parties, QKD state of the art has witnessed remarkable progress in the last years. Nonetheless, although theoretically secure, QKD is not implementation-secure and until now, the study of physical vulnerabilities in QKD setups has mainly focused on the optical channel. The concept of attacking a cryptographic system via its physical characteristics and associated leakages, known as side-channel analysis, was firstly introduced in classical cryptography, with the seminal work of Paul Kosher. Since then, power and electromagnetic side-channel analysis have become a staple in classical cryptanalysis. However, these concepts have hardly been applied to QKD. In this work, we propose and implement a new method for side-channel analysis on QKD systems, by exploiting the power consumption of the electronic driver controlling the electro-optical components of the QKD transmitter. For high-rate transmission, QKD modules typically require electronic drivers, such as Field Programmable Gate Arrays (FPGAs). Here, we will show that the FPGA's power consumption can leak information about the QKD operation, and consequently the transmitted key. The analysis was performed on the QKD transmitter at the University of Padua. Our results are consistent and show critical information leakage, having reached a maximum accuracy of 73.35% in the prediction of transmitted random keys at 100 MHz repetition frequency.

Random number generators underpin the security of current and future cryptographic systems and are therefore a likely target for attackers. Quantum random number generators have been hailed as the ultimate sources of randomness. However, as shown in this work, the susceptibility of the sensitive electronics required to implement such devices poses a serious threat to their security. We present the first out-of-band electromagnetic injection attack on a quantum random number generator through which an adversary can gain full control of the output. In our first experiment, the adversary forces the binary output of the generator to become an alternating string of 1s and 0s, with near 100% success. This attack may be spotted by a vigilant user performing statistical tests on their output strings. We therefore envisage a second more subtle attack in which the adversary forces the output to be a random pattern known to them, thus rendering any protection based on statistical tests ineffective.

We present two different QRNG devices which allow to maximize the entropy extraction of a system and so the generation rate. The two devices use single-photon avalanche diode along (SPAD) with FPGA device. The first device, Randy, uses only one SPAD and with a post-processing based on the Peres algorithm [Y. Peres, Ann. Statist. 20, 590 (1992)], has a generation rate of 1.8 Mbit/s. The second device, LinoSPAD, which is a CMOS SPAD array based device and integrate also a time-to-digital converter (TDC) on its FPGA, has a final generation rate of 310 Mbit/s thanks to a improved post-processing procedure which also includes the Zhou-Bruck algorithm [H. Zhou and J. Bruck, arXiv:1209.0726 (2012)].