Physical security of deep learning on edge devices: Comprehensive evaluation of fault injection attack vectors

2021 ◽  
Vol 120 ◽  
pp. 114116
Author(s):  
Xiaolu Hou ◽  
Jakub Breier ◽  
Dirmanto Jap ◽  
Lei Ma ◽  
Shivam Bhasin ◽  
...  
Keyword(s):  
Deep Learning ◽  
Comprehensive Evaluation ◽  
Fault Injection ◽  
Physical Security ◽  
Fault Injection Attack

scholarly journals Investigating the impact of pre-processing techniques and pre-trained word embeddings in detecting Arabic health information on social media

2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Yahya Albalawi ◽  
Jim Buckley ◽  
Nikola S. Nikolov
Keyword(s):  
Social Media ◽  
Deep Learning ◽  
Comprehensive Evaluation ◽  
Classification Problem ◽  
Data Sets ◽  
Word Embeddings ◽  
Data Set ◽  
Lower Accuracy ◽  
Health Related ◽  
The Impact

AbstractThis paper presents a comprehensive evaluation of data pre-processing and word embedding techniques in the context of Arabic document classification in the domain of health-related communication on social media. We evaluate 26 text pre-processings applied to Arabic tweets within the process of training a classifier to identify health-related tweets. For this task we use the (traditional) machine learning classifiers KNN, SVM, Multinomial NB and Logistic Regression. Furthermore, we report experimental results with the deep learning architectures BLSTM and CNN for the same text classification problem. Since word embeddings are more typically used as the input layer in deep networks, in the deep learning experiments we evaluate several state-of-the-art pre-trained word embeddings with the same text pre-processing applied. To achieve these goals, we use two data sets: one for both training and testing, and another for testing the generality of our models only. Our results point to the conclusion that only four out of the 26 pre-processings improve the classification accuracy significantly. For the first data set of Arabic tweets, we found that Mazajak CBOW pre-trained word embeddings as the input to a BLSTM deep network led to the most accurate classifier with F1 score of 89.7%. For the second data set, Mazajak Skip-Gram pre-trained word embeddings as the input to BLSTM led to the most accurate model with F1 score of 75.2% and accuracy of 90.7% compared to F1 score of 90.8% achieved by Mazajak CBOW for the same architecture but with lower accuracy of 70.89%. Our results also show that the performance of the best of the traditional classifier we trained is comparable to the deep learning methods on the first dataset, but significantly worse on the second dataset.

Vision-based Autonomous Vehicle Recognition

2021 ◽  
Vol 54 (4) ◽  
pp. 1-37
Author(s):  
Azzedine Boukerche ◽  
Xiren Ma
Keyword(s):  
Deep Learning ◽  
Intelligent Transportation Systems ◽  
Comprehensive Evaluation ◽  
Autonomous Vehicle ◽  
Transportation Systems ◽  
Regional Security ◽  
Future Research ◽  
Learning Abilities ◽  
Vehicle Recognition ◽  
Model Recognition

Vision-based Automated Vehicle Recognition (VAVR) has attracted considerable attention recently. Particularly given the reliance on emerging deep learning methods, which have powerful feature extraction and pattern learning abilities, vehicle recognition has made significant progress. VAVR is an essential part of Intelligent Transportation Systems. The VAVR system can fast and accurately locate a target vehicle, which significantly helps improve regional security. A comprehensive VAVR system contains three components: Vehicle Detection (VD), Vehicle Make and Model Recognition (VMMR), and Vehicle Re-identification (VRe-ID). These components perform coarse-to-fine recognition tasks in three steps. In this article, we conduct a thorough review and comparison of the state-of-the-art deep learning--based models proposed for VAVR. We present a detailed introduction to different vehicle recognition datasets used for a comprehensive evaluation of the proposed models. We also critically discuss the major challenges and future research trends involved in each task. Finally, we summarize the characteristics of the methods for each task. Our comprehensive model analysis will help researchers that are interested in VD, VMMR, and VRe-ID and provide them with possible directions to solve current challenges and further improve the performance and robustness of models.

scholarly journals A Review on Evaluation and Configuration of Fault Injection Attack Instruments to Design Attack Resistant MCU-Based IoT Applications

Electronics ◽  
2020 ◽  
Vol 9 (7) ◽  
pp. 1153
Author(s):  
Zahra Kazemi ◽  
David Hely ◽  
Mahdi Fazeli ◽  
Vincent Beroulle
Keyword(s):  
Fault Injection ◽  
Low Cost ◽  
Security And Privacy ◽  
Production Cycle ◽  
Fault Attacks ◽  
Primary Means ◽  
Level Of Knowledge ◽  
Iot Devices ◽  
Fault Injection Attack ◽  
Evaluation Platform

The Internet-of-Things (IoT) has gained significant importance in all aspects of daily life, and there are many areas of application for it. Despite the rate of expansion and the development of infrastructure, such systems also bring new concerns and challenges. Security and privacy are at the top of the list and must be carefully considered by designers and manufacturers. Not only do the devices need to be protected against software and network-based attacks, but proper attention must also be paid to recently emerging hardware-based attacks. However, low-cost unit software developers are not always sufficiently aware of existing vulnerabilities due to these kinds of attacks. To tackle the issue, various platforms are proposed to enable rapid and easy evaluation against physical attacks. Fault attacks are the noticeable type of physical attacks, in which the normal and secure behavior of the targeted devices is liable to be jeopardized. Indeed, such attacks can cause serious malfunctions in the underlying applications. Various studies have been conducted in other research works related to the different aspects of fault injection. Two of the primary means of fault attacks are clock and voltage fault injection. These attacks can be performed with a moderate level of knowledge, utilizing low-cost facilities to target IoT systems. In this paper, we explore the main parameters of the clock and voltage fault generators. This can help hardware security specialists to develop an open-source platform and to evaluate their design against such attacks. The principal concepts of both methods are studied for this purpose. Thereafter, we conclude our paper with the need for such an evaluation platform in the design and production cycle of embedded systems and IoT devices.

Sign in / Sign up

Export Citation Format

Share Document