Rowhammer is a hardware vulnerability in DRAM memory, where repeated access to hammer rows can induce bit flips in neighboring victim rows . Being a hardware vulnerability, rowhammer bypasses all the system memory protection, allowing adversaries to compromise the integrity and confidentiality of data. Rowhammer attacks have shown to enable privilege escalation, sandbox escape, and cryptographic key disclosures. A key requirement of all existing rowhammer attacks is that an attacker must have access to at least part of an exploitable hammer row. We term such rowhammer attacks as Explicit Hammer. Recently, several proposals suggest leveraging the spatial proximity between the accessed hammer rows and the location of the victim rows for a defense against rowhammer. These all aim to deny the attacker's permission to access hammer rows near sensitive data, thus defeating explicit hammer-based attacks. In this paper, we question the core assumption underlying these defenses. We present Implicit Hammer, a confused-deputy attack that causes accesses to hammer rows that the attacker is not allowed to access. It is a paradigm shift in rowhammer attacks since it crosses privilege boundary to stealthily rowhammer an inaccessible row by implicit DRAM accesses. Such accesses are achieved by abusing inherent features of modern hardware and/or software. We propose a generic model to rigorously formalize the necessary conditions to initiate implicit hammer and explicit hammer, respectively. Compared to explicit hammer, implicit hammer can defeat the advanced software-only defenses, stealthy in hiding itself and hard to be mitigated. To demonstrate the practicality of implicit hammer, we have created two implicit hammer's instances, called PThammer and SyscallHammer. Specifically, PThammer exploits the page-table-walk process of modern processors, inducing the processor to generate frequent accesses to protected memory locations. SyscallHammer exploits system call handler routine of OS kernels to implicitly access kernel memory.
30 October 2022 in International Journal of Electrical Power & Energy Systems
The backdoor attack on deep neural network models implants malicious data patterns in a model to induce attacker-desirable behaviors. Existing defense methods fall into the online and offline categories, in which the offline models achieve state-of-the-art detection rates but are restricted by heavy computation overhead. In contrast, their more deployable online counterparts lack the means to detect source-specific backdoors with large sizes. This work proposes a new online backdoor detection method—Reverse Backdoor Distillation (RBD) to handle issues associated with source-specific and source-agnostic backdoor attacks. RBD, designed with the novel perspective of distilling instead of erasing backdoor knowledge, is a complementary backdoor detection methodology that can be used in conjunction with other online backdoor defenses. Considering the fact that trigger data will cause overwhelming neuron activation while clean data will not, RBD distills backdoor attack pattern knowledge from a suspicious model to create a shadow model, which is subsequently deployed online along with the original model in scope to predict a backdoor attack. We extensively evaluate RBD on several datasets (MNIST, GTSRB, CIFAR-10) with diverse model architectures and trigger patterns. RBD outperforms online benchmarks in all experimental settings. Notably, RBD demonstrates superior capability in detecting source-specific attacks, where comparison methods fail, underscoring the effectiveness of our proposed technique. Moreover, RBD achieves a computational savings of at least 97%.
With the popularity of vision sensors and the advancement of image processing technology, machine vision tasks based on image matching have received widespread attention. Matching methods based on handcrafted features have strong interpretability, but they are susceptible to noise in complex environments. Additionally, deep learning has shown potential in improving matching performance, but at the cost of large amounts of specific samples and computing resources. In recent years, there have been efforts to explore how to combine handcrafted features and deep features to enhance performance, but most of them only consider their sequential relationship. To address these challenges, we propose a novel image matching scheme with a fine-grained decision level fusion method that fully considers the judgments of handcrafted features and deep features for potential correspondence points, effectively combining handcrafted and deep features in a novel way. We also train generic features on public datasets to handle unseen scenes. We use RootSIFT as prior knowledge to guide the extraction of deep features, significantly reducing computational overhead. Additionally, we carefully design preprocessing steps by incorporating device attitude information. We conduct image matching experiments on public datasets and our own captured dataset. Compared to the existing methods, as illustrated in our experimental results, the proposed scheme achieves an overall 2.5-6x more correct matches with improved robustness. Finally, we design a target localization algorithm incorporating an UAV to achieve accurate and efficient localization using the proposed matching scheme and weighted least squares estimation algorithm, demonstrating the versatility of the proposed image matching scheme.
Despite that much progress has been reported in gait recognition, most of these existing works adopt lateral-view parameters as gait features, which requires large area of data collection environment and limits the applications of gait recognition in real-world practice. In this paper, we adopt frontal-view walking sequences rather than lateral-view sequences and propose a new gait recognition method based on multi-modal feature representations and learning. Specifically, we characterize walking sequences with two different kinds of frontal-view gait features representations, including holistic silhouette and dense optical flow. Pedestrian regions extraction is achieved by an improved YOLOv7 algorithm called Gait-YOLO algorithm to eliminate the effects of background interference. Multi-modal fusion module (MFM) is proposed to explore the intrinsic connections between silhouette and dense optical flow features by using squeeze and excitation operations at the channel and spatial levels. Gait feature encoder is further used to extract global walking characteristics, enabling efficient multi-modal information fusion. To validate the efficacy of the proposed method, we conduct experiments on CASIA-B and OUMVLP gait databases and compare performance of our proposed method with other existing state-of-the-art gait recognition methods.
29 November 2024 in IEEE Internet of Things Journal
Millimeter-wave (mmWave) radar sensors show significant promise in noncontact human-computer interaction. Using air-writing as a substitute for conventional input devices, such as keyboards and mice has become a pivotal topic in contemporary research. In response to the absence of a dataset in existing studies about air-writing digits and the insufficient exploration of real-time recognition in edge devices, we propose a real-time air-writing digit recognition framework based on mmWave radar, termed mmDigit. Initially, we use mmWave radar equipped with frequency-modulated continuous wave (FMCW) technology to collect digital echo data and design a data processing pipeline to track and reconstruct digital trajectory images. These images are subsequently fed into a lightweight neural network, which is only 6.9K in parameter size, for exploring the images’ quality and the recognition and cross-user capabilities of small-scale air-writing datasets. To enhance mmDigit’s performance, we implement a transfer learning strategy to accommodate a broader range of digit writing styles and habits, achieving a recognition accuracy of 99.14% and a cross-user capability of 94.13%. Additionally, applying a knowledge distillation strategy enables the lightweight network to extract and learn deep-layer features, thereby improving the cross-user recognition accuracy to 96.22%.