Search NASA⌕ Search

SEARCH · Search NASA

Results for “Adversarial attacks”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

Cyber-Informed Engineering: The Need for a New Risk Informed and Design Methodology

Current engineering and risk management methodologies do not contain the foundational assumptions required to address the intelligent adversary’s capabilities in malevolent cyber attacks. Current methodologies focus on equipment failures or human error as initiating events for a hazard, while cyber attacks use the functionality of a trusted system to perform operations outside of the intended design and without the operator’s knowledge. These threats can by-pass or manipulate traditionally engineered safety barriers and present false information, invalidating the fundamental basis of a safety analysis. Cyber threats must be fundamentally analyzed from a completely new perspective where neither equipment nor human operation can be fully trusted. A new risk analysis and design methodology needs to be developed to address this rapidly evolving threatscape.

CCE↗
DOE OSTI2015

The Need for Cyber-Informed Engineering Expertise for Nuclear Research Reactors

Engineering disciplines may not currently understand or fully embrace cyber security aspects as they apply towards analysis, design, operation, and maintenance of nuclear research reactors. Research reactors include a wide range of diverse co-located facilities and designs necessary to meet specific operational research objectives. Because of the nature of research reactors (reduced thermal energy and fission product inventory), hazards and risks may not have received the same scrutiny as normally associated with power reactors. Similarly, security may not have been emphasized either. However, the lack of sound cybersecurity defenses may lead to both safety and security impacts. Risk management methodologies may not contain the foundational assumptions required to address the intelligent adversary’s capabilities in malevolent cyber attacks. Although most research reactors are old and may not have the same digital footprint as newer facilities, any digital instrument and control function must be considered as a potential attack platform that can lead to sabotage or theft of nuclear material, especially for some research reactors that store highly enriched uranium. This paper will provide a discussion about the need for cyber-informed engineering practices that include the entire engineering lifecycle. Cyber-informed engineering as referenced in this paper is the inclusion of cybersecurity aspects into the engineering process. A discussion will consider several attributes of this process evaluating the long-term goal of developing additional cyber safety basis analysis and trust principles. With a culture of free information sharing exchanges, and potentially a lack of security expertise, new risk analysis and design methodologies need to be developed to address this rapidly evolving (cyber) threatscape.

CCE↗
DOE OSTI2017

Nuclear power plant instrumentation and control cyber security common vector access leading to relational common cause failures - 340

Nuclear power plants and facilities have been implementing digital system upgrades into their previously analog systems for well over twenty years. New nuclear facilities' control, security, and emergency preparedness systems are almost exclusively built on digital architectures with a high degree of communication between the various systems that are often integrated together into a central control station to aid in operation or security of the facility. As digital systems become more widespread in nuclear facility control system architectures, cyber security related issues have become a significant concern to operators, regulators, governments, and other groups. Among the many concerns related to digital systems and cyber security is the area of common cause and common mode failures. This paper introduces, defines, and discusses some sources of common cause failure from a cyber security perspective: common vector access. This refers to specific access points that an adversary can exploit through a single attack sequence that have the potential to provide relational failures through common cause on multiple components, subsystems, systems, or plants. This paper will further discuss interconnected processes where these access points may exist, the importance of limiting or controlling these pinch points, and some methods of protecting common vector access points. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

SURVEILLANCE DETECTION FOR TRANSPORTATION OPERATIONS PERSONNEL TO PREVENT HIJACKING, THEFT, SABOTAGE, AND MALICIOUS SECURITY EVENTS DURING TRANSPORTING NUCLEAR MATERIAL

The secure transportation of high-consequence materials, including nuclear and radiological assets, is a critical global priority in the face of escalating terrorism, security threats, and violent protests targeting these operations. Effective surveillance detection—the ability to identify, assess, and respond to potential threats across a continuum of scenarios—is paramount in addressing these challenges. This paper outlines a phased, multi-tiered training program designed to strengthen the surveillance detection capabilities of organizations responsible for nuclear material transport. The proposed training program adopts a progressive approach, gradually increasing in technical complexity to provide participants with comprehensive knowledge and tools for implementing robust security strategies. It targets a wide spectrum of stakeholders, including competent authorities, regulators, inspectors, shippers, carriers, law enforcement, and emergency response personnel, equipping them to plan, evaluate, and safeguard nuclear material transportation effectively. Each phase of the program emphasizes distinct elements of the surveillance detection continuum and transport security, focusing on critical topics such as threat identification, adversary task timelines, protective methodologies, and attack mitigation strategies. The training framework is anchored in technical exchanges and scenario-driven courses that reflect real-world complexities and challenges. By addressing the surveillance detection continuum comprehensively—from early threat assessment to active countermeasures—the program reinforces global efforts to secure nuclear assets. It aligns with international security objectives and fosters a strong security culture within participating organizations, ensuring personnel are prepared to counter potential threats and maintain the safe, secure movement of these materials. Ultimately, this initiative aims to enhance preparedness, security, and response capabilities, supporting the global mission to safeguard high-consequence materials against evolving threats.

Zineddin, Dr. Z. [ORNL] (ORCID:0009000848740725)↗
DOE OSTI2024

Beyond the Bridge: Contention-Based Covert and Side Channel Attacks on Multi-GPU Interconnect

Recently, high-speed interconnects like NVLink have become a standard integration in modern multi-GPU systems, serving as a crucial bridge between CPUs and GPUs. This study highlights the vulnerability of multi-GPU systems to covert and side-channel attacks arising from congestion on these interconnects. An adversary, operating without special permissions, can extract private information about a victim's activities by monitoring NVLink congestion. Specifically, we leverage this insight to develop a covert channel attack across two GPUs, achieving a respectable bandwidth of 33 Kbps. Additionally, we introduce a side-channel attack that allows attackers to fingerprint applications on a remote GPU by probing the shared NVLink interconnect. We underscore the urgent nature of the threats posed by these attacks and advocate for heightened awareness within both industry and academia to implement multi-GPU interconnects with enhanced security measures.

Zhang, Yicheng↗
DOE OSTI2021

Nominal and adversarial synthetic PMU data for standard IEEE test systems

GridSTAGE (Spatio-Temporal Adversarial scenario GEneration) is a framework for the simulation of adversarial scenarios and the generation of multivariate spatio-temporal data in cyber-physical systems. GridSTAGE is developed based on Matlab and leverages Power System Toolbox (PST) where the evolution of the power network is governed by nonlinear differential equations. Using GridSTAGE, one can create several event scenarios that correspond to several operating states of the power network by enabling or disabling any of the following: faults, AGC control, PSS control, exciter control, load changes, generation changes, and different types of cyber-attacks. Standard IEEE bus system data is used to define the power system environment. GridSTAGE emulates the data from PMU and SCADA sensors. The rate of frequency and location of the sensors can be adjusted as well. Detailed instructions on generating data scenarios with different system topologies, attack characteristics, load characteristics, sensor configuration, control parameters are available in the Github repository - https://github.com/pnnl/GridSTAGE. There is no existing adversarial data-generation framework that can incorporate several attack characteristics and yield adversarial PMU data. The GridSTAGE framework currently supports simulation of False Data Injection attacks (such as a ramp, step, random, trapezoidal, multiplicative, replay, freezing) and Denial of Service attacks (such as time-delay, packet-loss) on PMU data. Furthermore, it supports generating spatio-temporal time-series data corresponding to several random load changes across the network or corresponding to several generation changes. A Koopman mode decomposition (KMD) based algorithm to detect and identify the false data attacks in real-time is proposed in https://ieeexplore.ieee.org/document/9303022. Machine learning-based predictive models are developed to capture the dynamics of the underlying power system with a high level of accuracy under various operating conditions for IEEE 68 bus system. The corresponding machine learning models are available at https://github.com/pnnl/grid_prediction.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2017

Temporal Cyber Attack Detection.

Rigorous characterization of the performance and generalization ability of cyber defense systems is extremely difficult, making it hard to gauge uncertainty, and thus, confidence. This difficulty largely stems from a lack of labeled attack data that fully explores the potential adversarial space. Currently, performance of cyber defense systems is typically evaluated in a qualitative manner by manually inspecting the results of the system on live data and adjusting as needed. Additionally, machine learning has shown promise in deriving models that automatically learn indicators of compromise that are more robust than analyst-derived detectors. However, to generate these models, most algorithms require large amounts of labeled data (i.e., examples of attacks). Algorithms that do not require annotated data to derive models are similarly at a disadvantage, because labeled data is still necessary when evaluating performance. In this work, we explore the use of temporal generative models to learn cyber attack graph representations and automatically generate data for experimentation and evaluation. Training and evaluating cyber systems and machine learning models requires significant, annotated data, which is typically collected and labeled by hand for one-off experiments. Automatically generating such data helps derive/evaluate detection models and ensures reproducibility of results. Experimentally, we demonstrate the efficacy of generative sequence analysis techniques on learning the structure of attack graphs, based on a realistic example. These derived models can then be used to generate more data. Additionally, we provide a roadmap for future research efforts in this area.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Artificial Intelligence, the Final Piece to the Counterforce Puzzle?

It is widely expected that artificial intelligence (AI) will disrupt and fundamentally transform many aspects of social, economic, and political life over the coming decades. As we are in the early stages of AI adoption in most fields, many discussions of its potential necessarily impact remain vague and superficial, including where national security is concerned. Nevertheless, some scholars have productively explored AI’s impact on the character of warfare, the balance of power among states, and human society itself. This article contributes to this growing body of literature by examining how this new technology might interact with one of the most prominent features of the post-World War II strategic environment: nuclear weapons. Specifically, it asks whether AI-driven improvements to intelligence, surveillance, and reconnaissance (ISR) capabilities, i.e., automated image recognition/computer vision, could enable an effective counterforce capability and thereby imperil nuclear deterrence and weaken first-strike stability. Since the advent of the nuclear age, analysts have been divided over the meaning of nuclear weapons for international politics. Proponents of the “nuclear revolution” argue that nuclear weapons, particularly a secure second-strike capability, which would allow a state to retaliate with nuclear force after suffering a first strike, satisfy a state’s security needs because they present a potential for unacceptable damage that would deter any potential challenger from even considering aggression. In their view, international politics in the shadow of nuclear weapons would be largely peaceful, competition would be relegated to the margins, and nuclear superiority would be meaningless. However, recent research suggests that this may not be the case. In fact, nuclear weapon states have sought to hold at risk the nuclear arsenals of their adversaries throughout the nuclear age. Still, no country has attacked another’s operational nuclear arsenal. That is because the requirements for a disarming strike are considerably higher than for damage limitation or retaliation against nuclear attack. A state trying to disarm its opponent would have to be certain to destroy, or at least disable, all of its adversary’s nuclear weapons. Even one operational nuclear warhead could inflict unacceptable damage in a retaliatory blow. The inability to be certain that a first strike could find and eliminate all enemy nuclear weapons has been the main roadblock for effective counterforce. As the number of battlefield sensors grows and ISR capabilities improve, however, this roadblock is crumbling. Some expect it to be fully overcome with the help of AI.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
DOE OSTI2013

Decision Analysis Methods For the Analysis of Nuclear Terrorism Threats with Imperfect Information

Protection of commercial nuclear fuel cycles against adversary groups who may attempt to steal weapons-usable material is an issue of interest to the U.S. Department of Energy, Office of Nuclear Energy. The wide array of potential nuclear fuel cycle configurations and the security measures that could be deployed to protect them pose a daunting system design problem. In general, there are billions of potential design configurations. Moreover, the design choices are not independent – the effectiveness of one design feature or security measure can affect the desirability of other design choices. To explore this issue, we develop a mixed-integer optimization problem that solves a complex decision tree to allocate security resources across a commercial nuclear fuel cycle configuration. A nominal UREX+ reprocessing facility is used to demonstrate how the model allocates resources across a particular fuel cycle facility and shows how the problem can be scaled up to include other facilities and design features in the fuel cycle. In our model, the defender first chooses a set of safeguards to implement in order to mitigate the risk of a terrorist theft from various unit operations in the reprocessing facility. The attacker observes the defender’s strategy choice and executes an attack to acquire material, including the possibility of declining to attack at all. The model considers the terrorist as an intelligent adversary with a set of attack strategies known to the defender. The strategic choices made by both the defender and attacker affect the probability of a successful attack by the terrorist and in turn affect the distribution of consequences to the defender. We use a decision tree to illustrate the interactions among costs, decisions, correlated uncertainties, and outcomes, and then show how to formulate the problem as a mixed-integer program with an equivalent solution. The novel features of our formulation include the following. First, as discussed above, we use a sequential game-theoretic formulation (a Stackelberg game). This formulation assumes the attacker can observe the defender’s actions before the attack is executed. Many previous game-theoretic studies have assumed attacker and defender act simultaneously without knowledge of the other’s strategy choice. We believe the sequential game formulation is more applicable for knowledgeable insider adversaries, the focus of this study. Second, we build on our previous work that has developed a methodology to reflect the correlation among probability distributions represented as discrete chance nodes in a decision tree. We illustrate how this methodology can be applied to two chance nodes in a decision tree that relate the quality of the materials that might be stolen or diverted by an adversary with the yield of a weapon that might be constructed from that material. In a decision tree, the dependence of the yield on the quality of the material could be modeled with conditional probabilities of the weapon yield given the quality of the materials obtained. For three outcome discretizations, this would require the assessment of nine conditional probabilities from domain experts as part of the model building effort. In contrast, our methodology will allow this same representation of the tree based on the assessment of the correlation between these two chance nodes by the experts, which we believe is a much easier and more reliable assessment task. The advantage of this approach increases as the number of dependent chance nodes grows in a decision tree representation of this problem. Third, we extend this approach to the approximation of the dependence between discrete binary events such as success or failure represented by binary chance nodes in a decision tree or in the special case of a probabilistic risk analysis (PRA) based on an event tree. This extension is a novel application in this domain, and uses Monte Carlo simulation to determine the appropriate conditional probabilities on the binary chance nodes when the number of correlated uncertainties is more than two. Fourth, we show how these decision trees or probability trees can be formulated and solved as equivalent integer programming (IP) problems. The advantages of the IP representation include the opportunity to use the computing power associated with powerful optimization solver algorithms available on high speed computers to solve very large versions of these trees, and the ability to include several such models of different facilities into a much larger IP with connecting constraints that represent budget restrictions, number of available inspection opportunities in a given time window, or other dependencies among these facilities. Policy makers can use the resulting model to inform their decisions about how to assess terrorism risk and to safeguard the commercial nuclear fuel cycle.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2016

Using Machine Learning in Adversarial Environments.

Intrusion/anomaly detection systems are among the first lines of cyber defense. Commonly, they either use signatures or machine learning (ML) to identify threats, but fail to account for sophisticated attackers trying to circumvent them. We propose to embed machine learning within a game theoretic framework that performs adversarial modeling, develops methods for optimizing operational response based on ML, and integrates the resulting optimization codebase into the existing ML infrastructure developed by the Hybrid LDRD. Our approach addresses three key shortcomings of ML in adversarial settings: 1) resulting classifiers are typically deterministic and, therefore, easy to reverse engineer; 2) ML approaches only address the prediction problem, but do not prescribe how one should operationalize predictions, nor account for operational costs and constraints; and 3) ML approaches do not model attackers’ response and can be circumvented by sophisticated adversaries. The principal novelty of our approach is to construct an optimization framework that blends ML, operational considerations, and a model predicting attackers reaction, with the goal of computing optimal moving target defense. One important challenge is to construct a realistic model of an adversary that is tractable, yet realistic. We aim to advance the science of attacker modeling by considering game-theoretic methods, and by engaging experimental subjects with red teaming experience in trying to actively circumvent an intrusion detection system, and learning a predictive model of such circumvention activities. In addition, we will generate metrics to test that a particular model of an adversary is consistent with available data.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Cyote-attack Chain Estimator

Attack Chain Estimator (ACE) Application Overview The Attack Chain Estimator (ACE) Application is a sophisticated tool designed for the ingestion, classification, sequencing, and enrichment of cybersecurity threat reports. This application leverages advanced machine learning models and extensive historical data to provide comprehensive insights into cyber threats, specifically targeting Industrial Control Systems (ICS). Purpose The primary functions of the ACE Application include: Ingestion of Cybersecurity Threat Reporting: Capable of ingesting text-based threat reports in markdown or text file format. Supports ingestion of structured data from other sources in STIX/JSON format. Classification of Report’s Text-Based Events: Utilizes a DeBERTa classifier, specifically trained on cybersecurity data, to map the events to MITRE ATT&CK for ICS Tactics and Techniques. Classification is performed using multiple Jupyter notebooks and machine learning workflows hosted as FastAPI microservices: regex_data deberta_base_35_train_hft_classifier_mlflow.ipynb hft_regex_classifier_mlflow.ipynb param_train_hft_classifier_mlflow.ipynb regex_tactic_tech.ipynb Ordering of Tactics, Techniques, and Observable Events: Sequences the identified tactics, techniques, and events to form a coherent attack chain. Enrichment with Historical Attack Chain Details: Enhances the attack chain with details from historical attacks using a Markov model developed from CyOTE Precursor Analysis Report data. The Markov model is available as a FastAPI endpoint for seamless integration. Enrichment with Adversary Emulation Capabilities Data: Integrates adversary emulation capabilities data using MITRE Caldera for OT adversary abilities UUIDs. Export of Output Files: Provides options to export the enriched attack chain in JSON or CSV formats. Routing of Output to Other Applications: Facilitates routing of output to various platforms and applications, including: Threat Intelligence Platforms COREII Scout for Threat Intelligence Analysis COREII Modeling and Simulation for Adversary Emulation Technical Description The ACE Application is an advanced cybersecurity tool designed to provide detailed threat analysis and sequence generation. It is built on a robust architecture that integrates natural language processing, machine learning, and historical data modeling. Key Components: Data Ingestion Module: Handles the input of threat reports and data from various formats, ensuring flexibility in data sources. Classification Engine: Employs DeBERTa-based classifiers hosted as FastAPI microservices to analyze and classify threat report events in accordance with the MITRE ATT&CK framework for ICS. Sequence Generator: Orders the classified events into a logical attack chain, providing clear insight into the sequence of tactics and techniques used in the threat. Enrichment Engine: Integrates historical data and adversary emulation capabilities to enhance the attack chain with valuable context and additional details. The historical data enrichment is powered by a Markov model, which is available as a FastAPI endpoint. Export and Routing Module: Facilitates the export of the enriched attack chain in multiple formats and routes the output to designated applications for further analysis or emulation.

Paul, Tony [Idaho National Laboratory (INL), Idaho↗
DOE OSTI2024

Sequence-Based Anomaly Detection in Critical Infrastructure Networks

United States critical infrastructure faces new cyber threats from adversarial nation-state actors in the form of malware-free attacks. Traditional cybersecurity techniques use rules-based methods to identify indicators of compromise on networks, often missing these sophisticated attacks. Our approach leverages multiple state of the art machine learning models in a pipeline to identify abnormal network events through sequential analysis. We combine both device and packet-level information into individual events to characterize anomalous network actions. The model is trained and tested on real network traffic from the Idaho National Lab High Performance Computing (HPC) with greater than 98% precision. It is capable of flagging malicious tactics used by adversaries in malware-free attacks, severe changes to the network, and abnormal user activity by network devices.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2020

Effects of Jacobian Matrix Regularization on the Detectability of Adversarial Samples

The well-known vulnerability of Deep Neural Networks to adversarial samples has led to a rapid cycle of increasingly sophisticated attack algorithms and proposed defenses. While most contemporary defenses have been shown to be vulnerable to carefully configured attacks, methods based on gradient regularization and out-of-distribution detection have attracted much interest recently by demonstrating higher resilience to a broad range of attack algorithms. However, no study has yet investigated the effect of combining these techniques. In this paper, we consider the effect of Jacobian matrix regularization on the detectability of adversarial samples on the CIFAR-10 image benchmark dataset. We find that regularization has a significant effect on detectability, and in some cases can make an undetectable attack on a baseline model detectable. In addition, we give evidence that regularization may mitigate the known weaknesses of detectors to high-confidence adversarial samples. The defenses we consider here are highly generalizable, and we believe they will be useful for further investigations to transfer machine learning robustness to other data domains.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
DOE OSTI2025

A distributed voltage inference framework for cyber-physical attacks detection and localization in active distribution grids

The transition to active distribution grids with real-time monitoring and control depends on the proliferation of advanced communication networks and devices. This paradigm shift towards a cyber-physical architecture also introduces new vulnerabilities for adversaries to exploit and launch sophisticated cyber-physical attacks targeting grid observability. Current research highlights the challenges in distinguishing attacks on voltage phasor or nodal injection measurements and isolating multi-source attack locations in a multiphase distribution grid. The attack detection and localization methods in literature face accuracy issues, applications across diverse attack scenarios, or scalability limits. Here, to bridge these gaps, this paper proposes a distributed Voltage Inference framework for real-time detection and localization of cyber-physical attacks, addressing scalability, adaptability, and accuracy challenges in state-of-the-art methods. The proposed methodology leverages the distributed nature of the Voltage Inference framework through a two-step process of prediction and correction, together with a tractable graph partitioning approach, providing a reliable solution to identify compromised measurement sources and facilitate isolation. Extensive testing on IEEE 13 and 123-node distribution feeders underscores the algorithm’s efficacy, enhancing the security and resilience of active distribution grids against evolving cyber threats. Additionally, Hardware-in-the-Loop (HIL) implementation validates the proposed strategy’s practical applicability in real-world scenarios.

active distribution grids↗
DOE OSTI2020

Cyber-physical Architecture For Automated Responses (cyphar) In Adversarial Ot Environments

The ability to react to a malicious attack starts with high fidelity recognition, and with that, an agile response to the attack. The current operational technology (OT) systems for a critical infrastructure may include an intrusion detection system (IDS), but the ability to adapt to an intrusion is a human initiated response. Orchestrators, which are coming of age in the financial sector and allow for levels of automated response, are not prevalent in the OT space. To evolve to such responses in the OT space, a tradeoff analysis is first needed, The tradeoff analysis, which evaluates the mitigation benefits of responses versus the physical affects that result, inform an automated response decision. The intent of this paper is to provide a formulation of a tradeoff analysis and its use in advancing automated, agile responses.

Rieger, CraigG.↗
DOE OSTI2017

Robust Structural Analysis and Design of Distributed Control Systems to Prevent Zero Dynamics Attacks

We consider the design and analysis of robust distributed control systems (DCSs) to ensure the detection of integrity attacks. DCSs are often managed by independent agents and are implemented using a diverse set of sensors and controllers. However, the heterogeneous nature of DCSs along with their scale leave such systems vulnerable to adversarial behavior. To mitigate this reality, we provide tools that allow operators to prevent zero dynamics attacks when as many as p agents and sensors are corrupted. Such a design ensures attack detectability in deterministic systems while removing the threat of a class of stealthy attacks in stochastic systems. To achieve this goal, we use graph theory to obtain necessary and sufficient conditions for the presence of zero dynamics attacks in terms of the structural interactions between agents and sensors. We then formulate and solve optimization problems which minimize communication networks while also ensuring a resource limited adversary cannot perform a zero dynamics attacks. Polynomial time algorithms for design and analysis are provided.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2013

Towards A Theory of Autonomous Reconstitution of Compromised Cyber-Systems

The ability to maintain mission-critical operations in cyber-systems in the face of disruptions is critical. Faults in cyber systems can come from accidental sources (e.g., natural failure of a component) or deliberate sources (e.g., an intelligent adversary). Natural and intentional manipulation of data, computing, or coordination are the most impactful ways that an attacker can prevent an infrastructure from realizing its mission goals. Under these conditions, the ability to reconstitute critical infrastructure becomes important. Specifically, the question is: Given an intelligent adversary, how can cyber systems respond to keep critical infrastructure operational? In cyber systems, the distributed nature of the system poses serious difficulties in maintaining operations, in part due to the fact that a centralized command and control apparatus is unlikely to provide a robust framework for resilience. Resilience in cyber-systems, in general, has several components, and requires the ability to anticipate and withstand attacks or faults, as well as recover from faults and evolve the system to improve future resilience. The recovery effort (and any subsequent evolution) may require significant reconfiguration of the system (at all levels – hardware, software, services, permissions, etc.) if the system is to be made resilient to further attack or faults. This is especially important in the case of ongoing attacks, where reconfiguration decisions must be taken with care to avoid further compromising the system while maintaining continuity of operations. Collectively, we will label this recovery and evolution process as “reconstitution”. Currently, reconstitution is performed manually, generally after-the-fact, and usually consists of either standing up redundant systems, check-points (rolling back the configuration to a “clean” state), or re-creating the system using “gold-standard” copies. For enterprise systems, such reconstitution may be performed either directly on hardware, or using virtual machines. A significant challenge within this context is the ability to verify that the reconstitution is performed in a manner that renders the cyber-system resilient to ongoing and future attacks or faults. Fundamentally, the need is to determine optimal configuration of the cyber system when a fault is determined to be present. While existing theories for fault tolerance (for example, Byzantine fault tolerance) can guarantee resilience under certain conditions, in practice, these theories can break down in the face of an intelligent adversary. Further, it is difficult, in a dynamically evolving environment, to determine whether the necessary conditions for resilience have been met, resulting in difficulties in achieving resilient operation. In addition, existing theories do not sufficiently take into account the cost for attack and defense (the adversary is generally assumed to have infinite resources and time), hierarchy of importance (all network resources are assumed to be equally important), and the dynamic nature of some attacks (i.e., as the attack evolves, can resilience be maintained?). Alternative approaches to resilience based on a centralized command and control structure suffer from a single-point-failure. This paper presents preliminary research towards concepts for effective autonomous reconstitution of compromised cyber systems. We describe a mathematical framework as a first step towards a theoretical basis for autonomous reconstitution in dynamic cyber-system environments. We then propose formulating autonomous reconstitution as an optimization problem and describe some of the challenges associated with this formulation. This is followed by a brief discussion on potential solutions to these challenges.

asymmetric resilience↗