Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “adversarial detection”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Inferring adversarial behaviour in cyber‐physical power systems using a Bayesian attack graph approach

Abstract Highly connected smart power systems are subject to increasing vulnerabilities and adversarial threats. Defenders need to proactively identify and defend new high‐risk access paths of cyber intruders that target grid resilience. However, cyber‐physical risk analysis and defense in power systems often requires making assumptions on adversary behaviour, and these assumptions can be wrong. Thus, this work examines the problem of inferring adversary behaviour in power systems to improve risk‐based defense and detection. To achieve this, a Bayesian approach for inference of the Cyber‐Adversarial Power System (Bayes‐CAPS) is proposed that uses Bayesian networks (BNs) to define and solve the inference problem of adversarial movement in the grid infrastructure towards targets of physical impact. Specifically, BNs are used to compute conditional probabilities to queries, such as the probability of observing an event given a set of alerts. Bayes‐CAPS builds initial Bayesian attack graphs for realistic power system cyber‐physical models. These models are adaptable using collected data from the system under study. Then, Bayes‐CAPS computes the posterior probabilities of the occurrence of a security breach event in power systems. Experiments are conducted that evaluate algorithms based on time complexity, accuracy and impact of evidence for different scales and densities of network. The performance is evaluated and compared for five realistic cyber‐physical power system models of increasing size and complexities ranging from 8 to 300 substations based on computation and accuracy impacts.

Sahu, Abhijeet↗
DOE OSTI2026

3D reconstruction and neural rendering for adversarial machine learning

While evasion attacks on computer vision systems have been widely studied, creating attacks that remain effective under significant changes in viewpoint continues to be challenging. Traditional approaches often rely on affine transformations of images, but these approaches degrade at larger perspective shifts and often produce unrealistic or ineffective perturbations. Recent methods use differentiable renderers to improve viewpoint robustness, but they typically depend on manually constructed 3D models. We introduce a semi-automated pipeline that generates physically printable and perspective-invariant adversarial patches using only a small set of 2D images. Our method integrates 3D reconstruction, neural rendering, adversarial patch optimization, and an object detection victim model into a unified workflow. We use 2D Gaussian Splatting for high fidelity mesh reconstruction and FlexPara for surface parameterization that produces texture maps suitable for patch editing. Together, these components form a fully differentiable pipeline in PyTorch3D that links texture modification to model outputs, enabling efficient optimization of patches that remain effective across many viewpoints. The complete process, from image capture to patch printing and physical evaluation, can be completed within a few hours. We demonstrate the effectiveness of the resulting patches through attacks on the YOLOv8 object detection model and discuss remaining challenges and opportunities for improving robustness and scalability.

Singhvi, Vivaan [ORNL] (ORCID:0009000586288221)↗
DOE OSTI2023

A Data Processing Pipeline for Socio-Technical Network Analysis [Slides]

With the rapid adoption of emerging technologies, there is a need to catalog and model sociotechnical interdependencies that have been historically used to influence the operation of Critical Infrastructure networks including the impacts of mergers and acquisitions, hostile takeovers, and foreign investment. Our research intends to address this need with two primary contributions. First, we have developed a data curation and processing pipeline to generate sociotechnical networks extracted from a variety of data sources including SEC filings and infrastructure asset databases. The pipeline, implemented in Apache Airflow, extracts and normalizes the representation of entities and relations, specified within ontologies. Second, networks produced by our pipeline enable the development of graph-theoretic metrics that consider the properties of network components in addition to its topology. Measures of network complexity, such as degree distribution, reachability analyses, temporal analysis, and community detection may be adapted to indicate adversarial organizational influence. Our intent is to provide an extensible, machine-actionable approach to quickly communicate such models, reproduce previous results, and adapt them to new, unanticipated situations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Bennett-Brassard 1984 quantum key distribution using conjugate homodyne detection

Optical homodyne detection has been widely used in continuous-variable (CV) quantum information processing for measuring field quadrature. In this paper we explore the possibility of operating a conjugate homodyne detection system in “photon counting” mode to implement discrete-variable (DV) quantum key distribution (QKD). A conjugate homodyne detection system, which consists of a beam splitter followed by two optical homodyne detectors, can simultaneously measure a pair of conjugate quadratures X and P of the incoming quantum state. In classical electrodynamics, X 2 + P 2 is proportional to the energy (the photon number) of the input light. In quantum optics, X and P do not commute and thus the above photon-number measurement is intrinsically noisy. This implies that a blind application of standard security proofs of QKD could result in pessimistic performance. We overcome this obstacle by taking advantage of two special features of the proposed detection scheme. First, the fundamental detection noise associated with vacuum fluctuations cannot be manipulated by an external adversary. Second, the ability to reconstruct the photon number distribution at the receiver's end can place additional constraints on possible attacks from the adversary. As an example, we study the security of the BB84 QKD using conjugate homodyne detection and evaluate its performance through numerical simulations. This study may open the door to a family of QKD protocols, complementary to the well-established DV-QKD based on single-photon detection and CV-QKD based on coherent detection.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Semantic Stealth: Crafting Covert Adversarial Patches for Sentiment Classifiers Using Large Language Models

Deep learning models have been shown to be vulnerable to adversarial attacks, in which perturbations to their inputs cause the model to produce incorrect predictions. As opposed to adversarial attacks in computer vision, where small changes introduced to pixel values can drastically alter a model's output while remaining imperceptible to humans, text-based attacks are difficult to conceal due to the discrete nature of tokens. Consequently, unconstrained gradient-based attacks often produce adversarial examples that lack semantic meaning, rendering them detectable through visual inspection or perplexity filters. In contrast to methods that rely on gradient-based optimization in the embedding space, we propose an approach that leverages a Large Language Model's ability to generate grammatically correct and semantically meaningful text to craft adversarial patches that seamlessly blend in with the original input text. These patches can be used to alter the behavior of a target model, such as a text classifier. Since our approach does not rely on gradient backpropagation, it only requires access to the target model's confidence scores, making it a grey-box attack. We demonstrate the feasibility of our approach using open-source LLMs, including Intel's Neural Chat, Llama2, and Mistral-Instruct, to generate adversarial patches capable of altering the predictions of a distilBERT model fine-tuned on the IMDB reviews dataset for sentiment classification.

Roa Carvajal, Maria↗
DOE OSTI2024

Radio Frequency Spectrum Audit to Inventory Private Cellular Base Station Infrastructure

The ever-changing cellular communication landscape makes it difficult to identify, map, and localize cellular base stations. Localizing cellular base stations provides various advantages, including information security, cybersecurity, spectrum management, and interference detection. For example, the MITRE ATT&CK® (Adversarial Tactics, Techniques, and Common Knowledge architecture) [1] and Common Attack Pattern Enumeration and Classification [2] emphasize the importance of being able to minimize the cyber security threat presented by unregulated private cellular base stations (PCBS). The majority of published research looks at the malicious use of PCBSs and focuses on using data retrieved from user equipment (UE), data obtained from an application on the UE, or data shared between the UE and a mobile network to locate it. This innovative strategy, however, focuses on the passively discovered uniqueness of radio frequency (RF) transmissions from commercial cellular infrastructure received in a designated monitoring position (DMP).

42 ENGINEERING↗
DOE OSTI2025

IoT Firmware Emulation and Its Security Application in Fuzzing: A Critical Revisit

As IoT devices with microcontroller (MCU)-based firmware become more common in our lives, memory corruption vulnerabilities in their firmware are increasingly targeted by adversaries. Fuzzing is a powerful method for detecting these vulnerabilities, but it poses unique challenges when applied to IoT devices. Direct fuzzing on these devices is inefficient, and recent efforts have shifted towards creating emulation environments for dynamic firmware testing. However, unlike traditional software, firmware interactions with peripherals that are significantly more diverse presents new challenges for achieving scalable full-system emulation and effective fuzzing. This paper reviews 27 state-of-the-art works in MCU-based firmware emulation and its applications in fuzzing. Instead of classifying existing techniques based on their capabilities and features, we first identify the fundamental challenges faced by firmware emulation and fuzzing. We then revisit recent studies, organizing them according to the specific challenges they address, and discussing how each specific challenge is addressed. We compare the emulation fidelity and bug detection capabilities of various techniques to clearly demonstrate their strengths and weaknesses, aiding users in selecting or combining tools to meet their needs. Finally, we highlight the remaining technical gaps and point out important future research directions in firmware emulation and fuzzing.

Zhou, Wei (ORCID:0000000178340839)↗
DOE OSTI2020

Cycle-Consistent Adversarial Networks for Realistic Pervasive Change Generation in Remote Sensing Imagery

This paper introduces a new method of generating realistic pervasive changes in the context of evaluating the effectiveness of change detection algorithms in controlled settings. The method - a cycle-consistent adversarial network (CycleGAN) - requires low quantities of training data to generate realistic changes. Here we show an application of CycleGAN in creating realistic snow-covered scenes of multispectral Sentinel-2 imagery, and demonstrate how these images can be used as a test bed for anomalous change detection algorithms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

On the Abuse and Detection of Polyglot Files

A polyglot is a file that is valid in two or more formats. Polyglot files pose a problem for file-upload and generative AI web interfaces that rely on format identification to determine how to securely handle incoming files. In this work we found that existing file-format and embedded-file detection tools, even those developed specifically for polyglot files, fail to reliably detect polyglot files used in the wild. To address this issue, we studied the use of polyglot files by malicious actors in the wild, finding 30 polyglot samples and 15 attack chains that leveraged polyglot files. Using knowledge from our survey of polyglot usage in the wild---the first of its kind---we created a novel data set based on adversary techniques. We then trained a machine learning detection solution, PolyConv, using this data set. PolyConv achieves a precision-recall area-under-curve score of 0.999 with an F1 score of 99.20% for polyglot detection and 99.47% for file-format identification, significantly outperforming all other tools tested. We developed a content disarmament and reconstruction tool, ImSan, that successfully sanitized 100% of the tested image-based polyglots, which were the most common type found via the survey. Our work provides concrete tools and suggestions to enable defenders to better defend themselves against polyglot files, as well as directions for future work to create more robust file specifications and methods of disarmament.

Oesch, T [ORNL] (ORCID:0000000269091022)↗
DOE OSTI2023

Proactive Intrusion Detection and Mitigation System

SAND2023-05661O The proactive intrusion detection and mitigation system (PIDMS) provides grid-edge situational awareness for cybersecurity defense by capturing real-time distributed energy resource (DER) network traffic and performance data with a novel approach that improves the detection and prevention of cyber-physical attacks. The PIDMS addresses the grid-edge security gap with real-time analysis of both network traffic and photovoltaic performance data to deliver a novel, cyber-physical intrusion detection system (IDS) approach that increases the accuracy and effectiveness of detection and mitigation. This hybrid IDS analysis enables dual monitoring that increases the workload of the adversary; both cyber and physical data would have to be simultaneously spoofed to evade detection. Furthermore, monitoring and analyzing cyber data are insufficient in some cases. For example, in an insider threat aimed at disrupting inverter grid-support functions where proper credentials and authentication are achieved, only the altered PV performance would indicate abnormal behavior. All in all, the PIDMS provides novel capabilities for: • Distributed, real-time cyber-physical detection and mitigation analysis • Cybersecurity defense for grid-edge systems • Analysis framework that can provide situational awareness across the transmission, distribution, and DER systems The PIDMS sensor is designed to collect cyber-physical data, process the data using machine-learning algorithms, detect abnormal events, and deploy mitigations. With these goals, the main functional PIDMS objectives are: • Capability to collect cyber-physical data • Onboard storage of cyber-physical data • Peer-to-peer communication • Computationally efficient machine-learning algorithms • Online cyber-physical data analysis • Alerting/visualization capabilities • Mitigation deployment capability with bump-in-the-wire (BITW) implementation Each of these functional objectives enable PIDMS to perform effective cyber-physical intrusion detection and mitigation. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Jones, Christian↗
DOE OSTI2022

Machine Learning-based False Data Injection Attack Detection and Localization in Power Grids

Cyberattacks on critical infrastructures can be catastrophic and bring nations to their knees. Therefore, detecting these attacks is crucial and challenging. This paper presents a novel approach for detecting and locating cyberattacks affecting an electrical power system. The adversary employs a man-in-the-middle technique to inject false data into the communication between distributed energy resources (DER) and Microgrid Controller (MGC) with the goal of disrupting power delivery. The approach for detection and localization is based on integrating multiple machine learning-based anomaly detection models that combine network traffic data and grid measurements. Experiments are performed to assess the method's performance using a hardware-in-the-loop real-time simulation testbed which includes Modbus TCP/IP communication. Power system topology and operating conditions are based on actual topology and real-world data provided by the Holy Cross Energy utility network. Results confirm that the method can be successfully employed for detecting and localizing cyberattacks.

Leao, Bruno P.↗
DOE OSTI2021

Transfer Learning using Denoising Auto-Encoders for Cellular-Level Annotation of Tumor in Pathology Slides

Adversarial examples can produce altered classifications using only seemingly innocuous, imperceptible perturbations to the original image. The imperceptibility of adversarial perturbations suggests that the corresponding classifiers use decision criteria different than those of a human. In a medical setting, inexplicable decision criteria confound a pathologist’s willingness to trust machine-generated annotations. Here, we analyze denoising tumor detection models to see if they are robust to imperceptible adversarial perturbations. Moreover, to be more fully trusted by pathologists, we require tumor detectors that generate interpretable annotations which segment pathology slides into tumorous and normal regions at the cellular level. We therefore compare transfer learning based on two different autoencoder architectures, one derived from a deep denoising bottleneck autoencoder and one from an over-complete sparse autoencoder. Both autoencoders were first trained in an unsupervised manner on a set of pathology slides drawn from the Camelyon16 dataset. The latent representations produced by each autoencoder were then passed to separate neural networks that were trained in a supervised manner on binary tumor-normal masks generated by pathologists at cellular resolution. Both tumor detectors supported better than 90% AUC PR as measured by the area under the precision/recall curve on a held-out pathology slide. To assess the underlying decision criteria used by both tumor detectors, we constructed imperceptible adversarial examples which reduced the AUC PR of both models to less than 70%. Random noise of the same amplitude had almost no effect on the AUC PR of either model. Additionally, each tumor detector was resistant to adversarial “transfer” attacks targeting the other. The adversarial perturbations showed strong characteristic differences: the deep denoising models perturbations were a very diffuse, seemingly unrecognizable pattern while the sparse coding models perturbations showed traces of tissue cells.

47 OTHER INSTRUMENTATION↗
DOE OSTI2021

SeismoGen: Seismic Waveform Synthesis Using GAN With Application to Seismic Data Augmentation

Abstract Detecting earthquake arrivals within seismic time series can be a challenging task. Visual, human detection has long been considered the gold standard but requires intensive manual labor that scales poorly to large data sets. In recent years, automatic detection methods based on machine learning have been developed to improve the accuracy and efficiency. However, the accuracy of those methods relies on access to a sufficient amount of high‐quality labeled training data, often tens of thousands of records or more. We aim to resolve this dilemma by answering two questions: (1) provided with a limited amount of reliable labeled data, can we use them to generate additional, realistic synthetic waveform data? and (2) can we use those synthetic data to further enrich the training set through data augmentation, thereby enhancing detection algorithms? To address these questions, we use a generative adversarial network (GAN), a type of machine learning model which has shown supreme capability in generating high‐quality synthetic samples in multiple domains. Once trained, our GAN model is capable of producing realistic seismic waveforms of multiple labels (noise and event classes). Applied to real Earth seismic data sets in Oklahoma, we show that data augmentation from our GAN‐generated synthetic waveforms can be used to improve earthquake detection algorithms in instances when only small amounts of labeled training data are available.

Wang, Tiantong↗
DOE OSTI2024

A Compound Data Poisoning Technique with Significant Adversarial Effects on Transformer-based Sentiment Classification Tasks

Transformer-based models have demonstrated much success in various natural language processing tasks. However, they are often vulnerable to adversarial attacks, such as data poisoning, which can intentionally fool the model into generating incorrect results. In this article, we present a novel, compound variant of a data poisoning attack on a transformer-based model that maximizes the poisoning effect while minimizing the scope of poisoning. Here we do so by combining the established data poisoning technique (label flipping) with a novel adversarial artifact selection and insertion technique aimed at minimizing detectability and the scope of the poisoning footprint. We find that by using a combination of these two techniques, we achieve a state-of-the-art attack success rate of approximately 90% while poisoning only 0.5% of the original training set, thus minimizing the scope and detectability of the poisoning action. These findings have the potential to advance the development of better data poisoning detection methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

DER Cybersecurity Detection and Response Suite

SAND2024-08475O The Distributed Energy Resource (DER) Cybersecurity Detection and Response Suite is a solution for distributed energy resource (DER) systems. The DER Security Orchestration, Automation, and Response (SOAR) solution that uses alerts from signature- and behavior-based Intrusion Detection Systems are intended to be deployed as bump-in-the-wire (BITW) devices in front of DER equipment. The fielded application would use multiple intrusion detection systems that report data to SOAR to respond to cyberattacks. The suite consists of two software components: • The proactive intrusion detection and mitigation system (PIDMS) secures grid-edge photovoltaic smart inverters and other equipment in distributed energy resource systems. It is a distributed BITW solution; cyber and physical data are automatically processed using network inspection tools and custom machine learning algorithms to detect abnormal events and correlate cyber-physical events. • The Security Orchestration, Automation, and Response for Distributed Energy Resources (SOAR4DER) application ingests data from several intrusion detection systems to quickly block attacks and revert DER systems to good states. Using a collection of intrusion detection system technologies on a BITW device, it incorporates physical and cyber data to detect abnormal and potential malicious behaviors. Multiple SOAR playbooks then use the intrusion detection system data streams to automatically defend the system. SOAR4DER system testing showed detection and response times under 30 seconds for all adversary reconnaissance, denial-of-service attacks, malicious Modbus commands, brute-force logins, and machine-in-the-middle attacks. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Johnson, Jay↗
NASA NTRS2019

Developing a Machine-Learning-Based Processing Framework for Twitter and Other Crowdsourced Data

Crowdsourced data streams such as Twitter and other social media are important sources of real-time and historical global information for Earth science applications. At the NASA Goddard Earth Sciences Data and Information Services Center (GES DISC), we have been exploring the Twitter data stream for its potential in augmenting the validation program of NASA's Global Precipitation Measurement (GPM) mission. To realize this potential, we need to increase the information density and enhance the quality of filtered precipitation tweets. We have implemented various components of a machine learning (ML)-based processing infrastructure for crowdsourced data that outputs, in this instance, useful and usable information derived from precipitation tweets. We have test enriched the Twitter stream with higher quality active tweets from those knowingly contributing to our effort and from existing crowdsourced programs (e.g., mPING, CoCoRaHS). We have experimented with various algorithms for processing tweets, including Naà ve Bayes, Convolutional Neural Network (CNN), Hierarchical Attention Network (HAN), and semi-supervised learning (with tri-training). Our current work focuses on (1) automated review of Earth science-related publications to determine relationships between discipline research needs and ML algorithms; (2) investigating Sequential Generative Adversarial Network (SeqGAN) for processing precipitation tweets for anomaly detection; and (3) managing crowdsourced data in a way that is compatible with existing NASA satellite data archives and using the data for ML applications. Key results include (1) network visualization of NLP-processed publications in various Earth science disciplines; (2) difference between GPM-linked, generated tweets and collected actual tweets that is small for GPM-determined light to moderate rain cases and high for GPM-determined heavy rain cases; and (3) identification of MongoDB for storing raw tweets and Zarr format for gridded tweets (compatible with GPM data). Our results have taken us a step closer to an operational ML-based tweet processing infrastructure and have already demonstrated that tweet-derived precipitation information is potentially useful for validation of Earth science satellite data.

Teng, William↗
DOE OSTI2021

Security proof of practical quantum key distribution with detection-efficiency mismatch

Quantum key distribution (QKD) protocols with threshold detectors are driving high-performance QKD demonstrations. The corresponding security proofs usually assume that all physical detectors have the same detection efficiency. However, the efficiencies of the detectors used in practice might show a mismatch depending on the manufacturing and setup of these detectors. A mismatch can also be induced as the different spatial-temporal modes of an incoming signal might couple differently to a detector. Here we develop a method that allows to provide security proofs without the usual assumption. Our method can take the detection-efficiency mismatch into account without having to restrict the attack strategy of the adversary. Especially, we do not rely on any photon-number cutoff of incoming signals such that our security proof is directly applicable to practical situations. We illustrate our method for a receiver that is designed for polarization encoding and is sensitive to a number of spatial-temporal modes. In our detector model, the absence of quantum interference between any pair of spatial-temporal modes is assumed. For a QKD protocol with this detector model, we can perform a security proof with characterized efficiency mismatch and without photon-number cutoff assumption. Our method also shows that in the absence of efficiency mismatch in our detector model, the key rate increases if the loss due to detection inefficiency is assumed to be outside of the adversary's control, as compared to the view where for a security proof this loss is attributed to the action of the adversary.

97 MATHEMATICS AND COMPUTING↗