Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “attack modeling”

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 991 records · Page 55
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Adaptive, self-tuning virtual sensing system for cyber-attack neutralization

An industrial asset may have a plurality of monitoring nodes, each monitoring node generating a series of monitoring node values over time representing current operation of the industrial asset. An abnormality detection computer may determine that an abnormal monitoring node is currently being attacked or experiencing a fault. An autonomous, resilient estimator may continuously execute an adaptive learning process to create or update virtual sensor models for that monitoring node. Responsive to an indication that a monitoring node is currently being attacked or experiencing a fault, a level of neutralization may be automatically determined. The autonomous, resilient estimator may then be dynamically reconfigured to estimate a series of virtual node values based on information from normal monitoring nodes, appropriate virtual sensor models, and the determined level of neutralization. The series of monitoring node values from the abnormal monitoring node or nodes may then be replaced with the virtual node values.

Abbaszadeh, Masoud↗
NASA NTRS1974

Transonic single-mode flutter and buffet of a low aspect ratio wing having a subsonic airfoil shape

Transonic flutter and buffet results obtained from wind-tunnel tests of a low aspect ratio semispan wing model are presented. The tests were conducted to investigate potential transonic aeroelastic problems of vehicles having subsonic airfoil sections. The model employed NACA 00XX-64 airfoil sections in the streamwise direction and had a 14 deg leading edge sweep angle. Aspect ratio, and average thickness were 4.0, 0.35, and 8 percent, respectively. The model was tested at Mach numbers from 0.6 to 0.95 at angles of attack from 0 deg to 15 deg. Two zero lift flutter conditions were found that involved essentially single normal mode vibrations. With boundary layer trips on the model, flutter occurred in a narrow Mach number range centered at about Mach 0.90. The frequency and motion of this flutter were like that of the first normal mode vibration. With the trips removed flutter occurred at a slightly high Mach number but in a mode strongly resembling that of the second normal mode.

Erickson, L. L.↗
DOE OSTI2026

Automated Membership Inference Attacks: Discovering MIA Signal Computations using LLM Agents

Membership inference attacks (MIAs), which enable adversaries to determine whether specific data points were part of a model's training dataset, have emerged as an important framework to understand, assess, and quantify the potential information leakage associated with machine learning systems. Designing effective MIAs is a challenging task that usually requires extensive manual exploration of model behaviors to identify potential vulnerabilities. In this paper, we introduce AutoMIA -- a novel framework that leverages large language model (LLM) agents to automate the design and implementation of new MIA signal computations. By utilizing LLM agents, we can systematically explore a vast space of potential attack strategies, enabling the discovery of novel strategies. Our experiments demonstrate AutoMIA can successfully discover new MIAs that are specifically tailored to user-configured target model and dataset, resulting in improvements of up to 0.18 in absolute AUC over existing MIAs. This work provides the first demonstration that LLM agents can serve as an effective and scalable paradigm for designing and implementing MIAs with SOTA performance, opening up new avenues for future exploration.

Tran, Toan Viet [Emory University]↗
DOE OSTI2026

Performing Numerical Analysis of Cybersecurity Options Using Dynamic Risk Analysis Tool EMRALD

Cyberattacks can have many different attack paths, durations, and goals. There are also many different mitigation options involving hardware, software, and/or humans. Considering a cyber threat should involve defense-in-depth methods and a quantitative or numerical evaluation of overall effectiveness against dynamic, time-dependent attacks to make cost and risk-informed decisions. Typical cyberattack modeling methods only provide a qualitative evaluation. The main areas of cybersecurity are confidentiality, integrity, and availability. For companies with cyber-physical systems such as advanced nuclear reactors, cyber-related safety is a requirement set by North American Electric Reliability and the U.S. Nuclear Regulatory Commission. They are also concerned about availability or reliability as a business case. As cyber threats are evolving to a business-for-hire structure, more attacks may focus on disrupting business success and reliability, causing financial and economic stability risk. Companies want to know business reliability and recovery from those threats, and that requires modeling physical behavior of the targets. Dynamic-state-based and Markov-based modeling provides a method for better cyber scenario modeling with different tools having issues such as state-base explosion. Dynamic modeling enables time and conditional features not found in other numerical evaluation methods. EMRALD (Event Modeling Risk Assessment using Lined Diagrams) is a dynamic risk analysis modeling and simulation tool and has features that reduce modeling issues. It has been used to model different time-dependent events including plant behavior and operator procedures. As a general modeling tool, EMRALD can also be used to model cyberattack scenarios with varying mitigation options and quantify effectiveness, producing numerical data for risk-informed decisions. This paper uses EMRALD to demonstrate that dynamic numerical risk analysis can be used for cyber threat modeling to provide insights for design decision-making and optimize defense strategies. Keywords: cyber modeling; cyber-physical systems; numerical cyber modeling

97 - MATHEMATICS AND COMPUTING↗
NASA NTRS2017

Design Guide for Aerodynamics Testing of Earth and Planetary Entry Vehicles in a Ballistic Range

The purpose of this manual is to aid in the design of an aerodynamics test of an earth or planetary entry capsule in a ballistic range. In this manual, much use is made of the results and experience gained in 50 years of ballistic range aerodynamics testing at the NASA Ames Research Center, and in particular, that gained in the last 27 years, while the author was working at NASA Ames. The topics treated herein include: Data to be obtained; flight data needed to design test; Reynolds number and dynamic similarity of flight trajectory and ballistic range test; capabilities of various ballistic ranges; Calculations of swerves due to average and oscillating lift and of drag-induced velocity decreases; Model and sabot design; materials, weights and stresses; Sabot separation; Launches at angle of attack and slapping with paper to produce pitch/yaw oscillations.

Test Design↗
NASA NTRS1984

Rotary balance data and analysis for the X-29A airplane for an angle-of-attack range of 0 deg to 90 deg

The rotational aerodynamic characteristics are discussed for a 1/8 scale model of the X-29A airplane. The effects of rotation on the aerodynamics of the basic model were determined, as well as the influence of airplane components, various control deflections, and several forebody modifications. These data were measured using a rotary balance, over an angle of attack range of 0 to 90 deg, for clockwise and counter clockwise rotations covering an omega b/2V range of 0 to 0.4.

Ralston, J. N.↗
DOE OSTI2021

Robustness of deep learning algorithms in astronomy -- galaxy morphology studies

Deep learning models are being increasingly adopted in wide array of scientific domains, especially to handle high-dimensionality and volume of the scientific data. However, these models tend to be brittle due to their complexity and overparametrization, especially to the inadvertent adversarial perturbations that can appear due to common image processing such as compression or blurring that are often seen with real scientific data. It is crucial to understand this brittleness and develop models robust to these adversarial perturbations. To this end, we study the effect of observational noise from the exposure time, as well as the worst case scenario of a one-pixel attack as a proxy for compression or telescope errors on performance of ResNet18 trained to distinguish between galaxies of different morphologies in LSST mock data. We also explore how domain adaptation techniques can help improve model robustness in case of this type of naturally occurring attacks and help scientists build more trustworthy and stable models.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS2000

Development of a Flush Airdata Sensing System on a Sharp-Nosed Vehicle for Flight at Mach 3 to 8

NASA Dryden Flight Research Center has developed a flush airdata sensing (FADS) system on a sharp-nosed, wedge-shaped vehicle. This paper details the design and calibration of a real-time angle-of-attack estimation scheme developed to meet the onboard airdata measurement requirements for a research vehicle equipped with a supersonic-combustion ramjet engine. The FADS system has been designed to perform in flights at Mach 3-8 and at -6 deg - 12 deg angle of attack. The description of the FADS architecture includes port layout, pneumatic design, and hardware integration. Predictive models of static and dynamic performance are compared with wind-tunnel results across the Mach and angle-of-attack range. Results indicate that static angle-of-attack accuracy and pneumatic lag can be adequately characterized and incorporated into a real-time algorithm.

Davis, Mark C.↗
NASA NTRS1966

Transonic aerodynamic investigation of tension shell and blunted 100 deg conical shapes for unmanned entry vehicles

An experimental investigation of a spherically blunted 100° cone and two tension shell configurations with varying nose-bluntness ratios for use in unmanned entry into low -density planetary atmospheres has been conducted by the National Aeronautics and Space Administration at Mach numbers from 0.30 to 1.20 and at angles of attack from approximately O° to 40°. The results show that the models were longitudinally stable, with centers of pressure located rearward of the base at all test conditions. The tension shell configurations exhibited substantially higher axial-force coefficients than the 100° cone although the variation of axial-force coefficient with Mach number at zero angle of attack was similar for all configurations. An increase in tension shell body length resulted in a reduction of the axial-force coefficients and a more forward center-of-pressure location. Variation in nose bluntness had no significant effect on the longitudinal aerodynamic characteristics of the tension shell configurations.

TRANSONIC SPEED↗
DOE OSTI2023

Machine Learning Based Resilience Testing of an Address Randomization Cyber Defense

Moving target defenses (MTDs) are widely used as an active defense strategy for thwarting cyberattacks on cyber-physical systems by increasing diversity of software and network paths. Recently, machine Learning (ML) and deep Learning (DL) models have been demonstrated to defeat some of the cyber defenses by learning attack detection patterns and defense strategies. It raises concerns about the susceptibility of MTD to ML and DL methods. Here, in this article, we analyze the effectiveness of ML and DL models when it comes to deciphering MTD methods and ultimately evade MTD-based protections in real-time systems. Specifically, we consider a MTD algorithm that periodically randomizes address assignments within the MIL-STD-1553 protocol—a military standard serial data bus. Two ML and DL-based tasks are performed on MIL-STD-1553 protocol to measure the effectiveness of the learning models in deciphering the MTD algorithm: 1) determining whether there is an address assignments change i.e., whether the given system employs a MTD protocol and if it does 2) predicting the future address assignments. The supervised learning models (random forest and k-nearest neighbors) effectively detected the address assignment changes and classified whether the given system is equipped with a specified MTD protocol. On the other hand, the unsupervised learning model (K-means) was significantly less effective. The DL model (long short-term memory) was able to predict the future addresses with varied effectiveness based on MTD algorithm's settings.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
NASA NTRS2006

Estimation of Longitudinal Unsteady Aerodynamics of a Wing-Tail Combination From Wind Tunnel Data

This paper presents an initial step toward model identification from wind tunnel data for an airliner configuration. Two approaches to modeling a transport configuration are considered and applied to both steady and large-amplitude forced-oscillation wind tunnel data taken over a wide range of angles of attack. Only limited conclusions could be drawn from this initial data set. Although model estimated time histories of normal force and pitching moment agree reasonably well with the corresponding measured values, model damping parameters did not, for some cases, have values consistent with small amplitude oscillatory data. In addition, large parameter standard errors implied poor information content for model structure determination and parameter estimation. Further investigation of the modeling problem for more general aerodynamic models is recommended with close attention to experiment design for obtaining parameters with high accuracy.

Murphy, Patrick C.↗
DOE OSTI2024

Double Visual Defense

This is the official code for the paper "Double Visual Defense: Adversarial Pre-training and Instruction Tuning for Improving Vision-Language Model Robustness". This code can be used to produce vision language models (VLMs), like LLaVA, with enhanced robustness to adversarial attacks (e.g. jailbreaks).

Bartoldson, Brian [Lawrence Livermore National Lab↗
NASA NTRS1971

A numerical method for computing three dimensional viscous supersonic flow fields about slender bodies

A numerical method is presented for the calculation of steady, three-dimensional, viscous, compressible flow fields about slender bodies at angle of attack and at supersonic speeds. Approximations are introduced in modeling the flow in the longitudinal direction. Accordingly, the flow fields calculated with the program were computed with a model that permits viscous crossflow together with inviscid axial flow. An analysis of the errors introduced by such a treatment is presented. Numerical calculations were made and compared with experimental results for an ogive-cylinder and an airplane fuselage configuration. Generally, good agreement with experiment was obtained. However, boundary layer separation and body vortex positions differed from experimental locations on the ogive-cylinder, and the shock induced by the fuselage canopy was predicted at a slightly different location.

Walitt, L.↗
NASA NTRS1974

Effect of external tank nose shape on the Rockwell International space shuttle vehicle 3, integrated configuration (IA37B)

Tests of several tank nose shapes were conducted as a part of the investigation of configuration changes to reduce drag for the integrated vehicle. The effect on the integrated vehicle aerodynamic characteristics of several tank nose shapes was investigated. The tank nose shapes investigated were the 600 inch (baseline) and 1204 inch radius ogives, and the 600 inch ogive plus a spike 360 inches long and 12.0 inches in diameter. Data were obtained over a Mach number range of 0.6 through 4.96 and for angles-of-attack and sideslip from -10 through +10 degrees. The model used was the 0.004-scale integrated vehicle model number 34-OTS.

Allen, E. C.↗
NASA NTRS1977

Force testing manual for the Langley 20-inch Mach 6 tunnel

Data reduction and procedures for conducting force tests in a 20 inch Mach 6 tunnel are described. A discussion of pretest and testing phases are included. Items that are to be checked during model design and construction are outlined as well as safety requirements, starting loads tests, instructions for data acquisition and model installation. Measurement of balance and model misalignment and instructions for calibrating the angle of attack screen are covered. Procedures for making reference pressure, attitude tare, and data runs are included. The 20 inch tunnel force program is examined, and a description of data recording system input and load contrast sheets is given. An appendix presents a description, operating characteristics, and Mach number calibration of the tunnel, as well as tunnel characteristics.

Keyes, J. W.↗
NASA NTRS1977

Aerodynamics of slender finned bodies at large angles of attack

In certain missions finned missiles perform slewing maneuvers. Here, large angles of attack are attained. Experimental data needed to understand the aerodynamics of such vehicles are presented. The purpose of this investigation was to study the interaction of the body flow field with that produced by the fins and the resulting effects on the aerodynamic forces and moments. The experiments were conducted at a nominal Mach number of 2.7 and angles of attack from 0 to 50 deg, with two different models. The tests were performed in a range of Reynolds number from 1.5 x 10 to the 6th to 4 x 10 to the 7th per foot (to cover both the laminar and fully turbulent regimes.) Several fin roll angles were investigated. Static pressures on both body and fin surfaces are reported.

Agnone, A. M.↗
NASA NTRS1978

Visualization of the separation and subsequent transition near the leading edge of airfoils

A visual study was performed using the low speed smoke wind tunnels with the objective of obtaining a better understanding of the structure of leading edge separation bubbles on airfoils. The location of separation, transition and reattachment for a cylindrical nose constant-thickness airfoil model were obtained from smoke photographs and surface oil flow techniques. These data, together with static pressure distributions along the leading edge and upper surface of the model, produced the influence of Reynolds number, angle of attack, and trailing edge flap angle on the size and characteristics of the bubble. Additional visual insight into the unsteady nature of the separation bubble was provided by high speed 16 mm movies. The 8 mm color movies taken of the surface oil flow supported the findings of the high speed movies and clearly showed the formation of a scalloped spanwise separation line at the higher Reynolds number.

Arena, A. V.↗
NASA NTRS1979

Pressure distributions on three different cruciform aft-control surfaces of a wingless missile at Mach 1.60, 2.36, and 3.70. Volume 2: Clipped delta tail

Pressure coefficients were obtained in the Langley Unitary Plan wind tunnel for a wingless missile with a clipped delta tail. The angle of attack was varied from -4 deg to 20 deg, model roll angle was varied from 0 deg to 90 deg in 22.5 deg increments, and tail deflections were 0 deg to - 15 deg. The pressures were measured on two adjacent tail surfaces using 91 pressure orifices per tail surface. Results are presented in plotted and tabular form.

Lamb, M.↗