Search

Cyber resilience is the ability of a system to resist and recover from a cyber attack, thereby restoring the system's functionality. Effective design and development of a cyber resilient system requires experimental methods and tools for quantitative measuring of cyber resilience. This paper describes an experimental method and test bed for obtaining resilience-relevant data as a system (in our case -- a truck) traverses its route, in repeatable, systematic experiments. We model a truck equipped with an autonomous cyber-defense system and which also includes inherent physical resilience features. When attacked by malware, this ensemble of cyber-physical features (i.e., "bonware") strives to resist and recover from the performance degradation caused by the malware's attack. We propose parsimonious mathematical models to aid in quantifying systems' resilience to cyber attacks. Using the models, we identify quantitative characteristics obtainable from experimental data, and show that these characteristics can serve as useful quantitative measures of cyber resilience., Comment: arXiv admin note: text overlap with arXiv:2302.04413, arXiv:2302.07941

The vulnerability of cyber-physical systems to cyber attack is well known, and the requirement to build cyber resilience into these systems has been firmly established. The key challenge this paper addresses is that maturing this discipline requires the development of techniques, tools, and processes for objectively, rigorously, and quantitatively measuring the attributes of cyber resilience. Researchers and program managers need to be able to determine if the implementation of a resilience solution actually increases the resilience of the system. In previous work, a table top exercise was conducted using a notional heavy vehicle on a fictitious military mission while under a cyber attack. While this exercise provided some useful data, more and higher fidelity data is required to refine the measurement methodology. This paper details the efforts made to construct a cost-effective experimentation infrastructure to provide such data. It also presents a case study using some of the data generated by the infrastructure., Comment: 6 pages, 2022 IEEE Military Communications Conference, pp. 855-860

We model a vehicle equipped with an autonomous cyber-defense system in addition to its inherent physical resilience features. When attacked, this ensemble of cyber-physical features (i.e., ``bonware'') strives to resist and recover from the performance degradation caused by the malware's attack. We model the underlying differential equations governing such attacks for piecewise linear characterizations of malware and bonware, develop a discrete time stochastic model, and show that averages of instantiations of the stochastic model approximate solutions to the continuous differential equation. We develop a theory and methodology for approximating the parameters associated with these equations., Comment: 6 pages, Invited Session on "Estimation and Learning in Stochastic Systems" for the 57th Annual Conference on Information Sciences and Systems. Co-sponsorship of Johns Hopkins University and the IEEE Information Theory Society

We identify quantitative characteristics of responses to cyber compromises that can be learned from repeatable, systematic experiments. We model a vehicle equipped with an autonomous cyber-defense system and which also has some inherent physical resilience features. When attacked by malware, this ensemble of cyber-physical features (i.e., "bonware") strives to resist and recover from the performance degradation caused by the malware's attack. We propose parsimonious continuous models, and develop stochastic models to aid in quantifying systems' resilience to cyber attacks., Comment: 7 pages, 2022 IEEE Military Communications Conference

Network intrusion detection systems (NIDS) are an essential defense for computer networks and the hosts within them. Machine learning (ML) nowadays predominantly serves as the basis for NIDS decision making, where models are tuned to reduce false alarms, increase detection rates, and detect known and unknown attacks. At the same time, ML models have been found to be vulnerable to adversarial examples that undermine the downstream task. In this work, we ask the practical question of whether real-world ML-based NIDS can be circumvented by crafted adversarial flows, and if so, how can they be created. We develop the generative adversarial network (GAN)-based attack algorithm NIDSGAN and evaluate its effectiveness against realistic ML-based NIDS. Two main challenges arise for generating adversarial network traffic flows: (1) the network features must obey the constraints of the domain (i.e., represent realistic network behavior), and (2) the adversary must learn the decision behavior of the target NIDS without knowing its model internals (e.g., architecture and meta-parameters) and training data. Despite these challenges, the NIDSGAN algorithm generates highly realistic adversarial traffic flows that evade ML-based NIDS. We evaluate our attack algorithm against two state-of-the-art DNN-based NIDS in whitebox, blackbox, and restricted-blackbox threat models and achieve success rates which are on average 99%, 85%, and 70%, respectively. We also show that our attack algorithm can evade NIDS based on classical ML models including logistic regression, SVM, decision trees and KNNs, with a success rate of 70% on average. Our results demonstrate that deploying ML-based NIDS without careful defensive strategies against adversarial flows may (and arguably likely will) lead to future compromises.

Machine learning is vulnerable to adversarial examples-inputs designed to cause models to perform poorly. However, it is unclear if adversarial examples represent realistic inputs in the modeled domains. Diverse domains such as networks and phishing have domain constraints-complex relationships between features that an adversary must satisfy for an attack to be realized (in addition to any adversary-specific goals). In this paper, we explore how domain constraints limit adversarial capabilities and how adversaries can adapt their strategies to create realistic (constraint-compliant) examples. In this, we develop techniques to learn domain constraints from data, and show how the learned constraints can be integrated into the adversarial crafting process. We evaluate the efficacy of our approach in network intrusion and phishing datasets and find: (1) up to 82% of adversarial examples produced by state-of-the-art crafting algorithms violate domain constraints, (2) domain constraints are robust to adversarial examples; enforcing constraints yields an increase in model accuracy by up to 34%. We observe not only that adversaries must alter inputs to satisfy domain constraints, but that these constraints make the generation of valid adversarial examples far more challenging., Comment: Accepted to the 28th ACM Conference on Computer and Communications Security. Seoul, South Korea