Abstract
During the last decade, Deep Neural Networks (DNN) have progressively been integrated on all types of platforms, from data centers to embedded systems including low-power processors and, recently, FPGAs. Neural Networks (NN) are expected to become ubiquitous in IoT systems by transforming all sorts of real-world applications, including applications in the safety-critical and security-sensitive domains. However, the underlying hardware security vulnerabilities of embedded NN implementations remain unaddressed. In particular, embedded DNN implementations are vulnerable to Side-Channel Analysis (SCA) attacks, which are especially important in the IoT and edge computing contexts where an attacker can usually gain physical access to the targeted device. A research field has therefore emerged and is rapidly growing in terms of the use of SCA including timing, electromagnetic attacks and power attacks to target NN embedded implementations. Since 2018, research papers have shown that SCA enables an attacker to recover inference models architectures and parameters, to expose industrial IP and endangers data confidentiality and privacy. Without a complete review of this emerging field in the literature so far, this paper surveys state-of-the-art physical SCA attacks relative to the implementation of embedded DNNs on micro-controllers and FPGAs in order to provide a thorough analysis on the current landscape. It provides a taxonomy and a detailed classification of current attacks. It first discusses mitigation techniques and then provides insights for future research leads.
Highlights
Accepted: 21 July 2021Every electronic device generates observable parasite signals during data computation that can leak internal information
After proposing a taxonomy of Side-Channel Analysis (SCA) attacks, we provide a detailed classification of current attacks according to several features such as target model and Deep Neural Networks (DNN) implementation, source of leakage, aim of the attack and the considered threat model
In this study we focus on passive SCA attacks on DNN implementations
Summary
Every electronic device generates observable parasite signals during (i.e., as an unintentional side effect of) data computation that can leak internal information. The sources of this information leakage, which are known as side-channels, can be computation timing, power consumption or electromagnetic (EM) emanation among others. By statistically analysing the measured side-channel data for each internal state together with a hypothesis on the processed data, an attacker can deduce some intermediate computation states This in turn allows the extraction of information about the processed data that would be otherwise inaccessible. The reader is referred to [9] for more detailed information with respect to DNNs. An MLP is a feed-forward network which is composed of several layers and each layer comprises a number of neurons. MLPs include at least three different layers: an input layer, an output layer and one or more hidden layers
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
