Search

Physical unclonable functions (PUFs) have been adopted in many resource-constrained Internet of Things (IoT) applications to provide effective and lightweight solutions for device authentication. However, an attacker can collect challenge–response pairs (CRPs) of a strong PUF, to build a machine learning (ML) model and mimic its behavior, i.e., predicting the responses of unseen challenges with high accuracy. Although several PUFs have been proposed to resist such modeling attacks, they incur high hardware overhead. Developing a PUF primitive with low hardware cost and high resistance to ML attacks is thus a crucial task. In this article, we propose the first response–feedback-based lightweight anti-ML-attack PUF (FLAM-PUF). It is only composed of one arbiter PUF (APUF) and one Galois linear-feedback shift register (LFSR), with some basic logic gates, reducing more than 62% hardware cost compared with the state-of-the-art robust strong PUFs. Specifically, FLAM-PUF leverages a cost-effective feedback loop structure to dynamically control and update the LFSR configuration. FLAM-PUF has two main characteristics: 1) it feeds back a 1-bit response in every cycle to intentionally poison the data of the CRP set for training. To resist ML-based modeling attacks, the 1-bit response can randomly update one coefficient of the feedback polynomial to implant more complex correlations into the model built by attackers and 2) it takes advantage of an $n-$ bit response feedback-controlled reconfigurable Galois LFSR to enlarge the original challenge space of the APUF. Extensive experimental results show that the proposed FLAM-PUF achieves near-optimal uniformity, uniqueness, and reliability. Our scheme works well under standard attack models with public crucial initial information. In particular, the prediction accuracy of modeling attacks against FLAM-PUF is nearly 50% under the four widely used ML algorithms, i.e., support vector machines (SVMs), logistic regression (LR), covariance matrix adaptation evolution strategy (CMA-ES), and deep neural networks (DNNs), indicating excellent resistance against these ML attacks. [ABSTRACT FROM AUTHOR]

Although deep neural networks (DNNs) have performed well on many perceptual tasks, they are vulnerable to adversarial examples that are generated by adding slight but maliciously crafted perturbations to benign images. Adversarial detection is an important technique for identifying adversarial examples before they are entered into target DNNs. Previous studies that were performed to detect adversarial examples either targeted specific attacks or required expensive computation. Designing a lightweight unsupervised detector is still a challenging problem. In this paper, we propose an A uto E ncoder-based A dversarial E xamples (AEAE) detector that can guard DNN models by detecting adversarial examples with low computation in an unsupervised manner. The AEAE includes only a shallow autoencoder that performs two roles. First, a well-trained autoencoder has learned the manifold of benign examples. This autoencoder can produce a large reconstruction error for adversarial images with large perturbations, so we can detect significantly perturbed adversarial examples based on the reconstruction error. Second, the autoencoder can filter out small noises and change the DNN's prediction on adversarial examples with small perturbations. It helps to detect slightly perturbed adversarial examples based on the prediction distance. To cover these two cases, we utilize the reconstruction error and prediction distance from benign images to construct a two-tuple feature set and train an adversarial detector using the isolation forest algorithm. We show empirically that AEAE is an unsupervised and inexpensive detector against most state-of-the-art attacks. Through the detection in these two cases, there is nowhere to hide adversarial examples. • We observe that adversarial detection is sensitive to the perturbation level. • We train a shallow autoencoder to find two key features from adversarial examples. • We propose a lightweight and unsupervised adversarial detector. [ABSTRACT FROM AUTHOR]

Input Method Editor (IME) is an indispensable component on current smartphones. With its assistance, the number of key presses is reduced, and non-Latin characters could be inputted. Furthermore, modern IMEs integrate several personalized features like reordering suggestion lists and predicting the next words based on user's input history. Such optimization improves the user experience but turns the IME dictionary into a pool of user privacy. Previous works have discussed the privacy risks coming from malicious IMEs. Indeed, they could cause security and privacy issues if installed by common users, but their impact is limited as the majority of IMEs are well-behaved. However, whether legitimate IMEs are bullet-proof is not answered before. In this paper, we make the first attempt to study the security implications of IME personalization and the back-end infrastructure on Android devices. In the end, we identify a critical vulnerability lying under the Android KeyEvent processing framework, which can be exploited to launch cross-app KeyEvent injection (CAKI) attack and bypass the app-isolation mechanism. By abusing such design flaw, an adversary can harvest entries from the personalized user dictionary of IME through an ostensibly innocuous app only asking for common permissions. Our evaluation over a broad spectrum of Android OSes, devices, and IMEs suggests such issue should be fixed immediately. All Android versions we examined (from very old 2.3.4 to the latest 6.0.1) and most IME apps we surveyed (11 out of 18) are vulnerable. User's private information, like contact names, location, etc., can be easily exfiltrated. Up to hundreds of millions of mobile users are under this threat. To mitigate this security issue, we propose a practical defense mechanism which augments the existing KeyEvent processing framework without forcing any change to IME apps. [ABSTRACT FROM AUTHOR]