As deep learning systems are widely adopted in safety-and securitycritical applications, such as autonomous vehicles, banking systems, etc., malicious faults and attacks become a tremendous concern, which potentially could lead to catastrophic consequences. In this paper, we initiate the first study of leveraging physical fault injection attacks on Deep Neural Networks (DNNs), by using laser injection technique on embedded systems. In particular, our exploratory study targets four widely used activation functions in DNNs development, that are the general main building block of DNNs that creates non-linear behaviors -ReLu, softmax, sigmoid, and tanh. Our results show that by targeting these functions, it is possible to achieve a misclassification by injecting faults into the hidden layer of the network. Such result can have practical implications for realworld applications, where faults can be introduced by simpler means (such as altering the supply voltage).

No abstract

LCD (linear complimentary dual) codes are linear codes that trivially intersect their duals. We address the question of an equivalent concept for lattices. We observe basic properties of the intersection of a lattice with its dual, and consider the construction of lattices from LCD codes using Construction A. Lattices obtained from the intersection of a code with its dual via Construction A are further discussed.

Neural networks have been shown to be vulnerable against fault injection attacks. These attacks change the physical behavior of the device during the computation, resulting in a change of value that is currently being computed. They can be realized by various fault injection techniques, ranging from clock/voltage glitching to application of lasers to rowhammer. In this paper we explore the possibility to reverse engineer neural networks with the usage of fault attacks. SNIFF stands for sign bit flip fault, which enables the reverse engineering by changing the sign of intermediate values. We develop the first exact extraction method on deep-layer feature extractor networks that provably allows the recovery of the model parameters. Our experiments with Keras library show that the precision error for the parameter recovery for the tested networks is less than 10 −13 with the usage of 64-bit floats, which improves the current state of the art by 6 orders of magnitude. Additionally, we discuss the protection techniques against fault injection attacks that can be applied to enhance the fault resistance.

The secrecy gain of a lattice is a lattice invariant used to characterize wiretap lattice codes for Gaussian channels. The secrecy gain has been classified for unimodular lattices up to dimension 23, and so far, a few sparse examples are known for lmodular lattices, with l = 2, 3. We propose some constructions of 5-modular lattices via the Construction A of lattices from linear codes, and study the secrecy gain of the resulting lattices.

Cryptographic implementations are often vulnerable against physical attacks, fault injection analysis being among the most popular techniques. On par with development of attacks, the area of countermeasures is advancing rapidly, utilizing both hardware-and software-based approaches. When it comes to software encoding countermeasures for fault protection and their evaluation, there are very few proposals so far, mostly focusing on single operations rather than cipher as a whole. In this paper we propose an evaluation framework that can be used for analyzing the effectivity of software encoding countermeasures against fault attacks. We first formalize the encoding schemes in software, helping us to define what properties are required when designing a fault protection. Based on these findings, we develop an evaluation metric that can be used universally to determine the robustness of a software encoding scheme against bit flip faults and instruction skips. We provide a way to select a code according to user criteria and also a dynamic code analysis method to estimate the level of protection of assembly implementations using encoding schemes. Finally, we verify our findings by implementing a block cipher PRESENT, protected by encoding scheme based on anticodes, and provide a detailed evaluation of this implementation using different codes.

Over the past decades, fault injection attacks have been extensively studied due to their capability to efficiently break cryptographic implementations. Fault injection attack models are normally determined by analyzing the cipher structure and finding exploitable spots in non-linear and permutation layers. However, this level of abstraction is often too high to distinguish vulnerable parts of software implementations, due to specific operations and optimizations. On the other hand, manually analyzing the assembly code requires non-negligible amount of time and expertise. In this paper, we propose an automated approach for analyzing cipher implementations in assembly. We represent the whole assembly program as a data flow graph so that the vulnerable spots can be found efficiently. Fault propagation is analyzed in a subgraph constructed from each vulnerable spot, allowing equations for Differential Fault Analysis (DFA) to be automatically generated. We have created a tool that implements our approach: DATAC – DFA Automation Tool for Assembly Code. We have successfully used this tool for attacking PRESENT- 80, being able to find implementation-specific vulnerabilities that can be exploited in order to recover the last round key with 16 faults. Our results show that DATAC is useful in finding attack spots that are not visible from the cipher structure, but can be easily exploited when dealing with real-world implementations.

No abstract