Biblio
Cyber-physical systems (CPS) can benefit by the use of learning enabled components (LECs) such as deep neural networks (DNNs) for perception and decision making tasks. However, DNNs are typically non-transparent making reasoning about their predictions very difficult, and hence their application to safety-critical systems is very challenging. LECs could be integrated easier into CPS if their predictions could be complemented with a confidence measure that quantifies how much we trust their output. The paper presents an approach for computing confidence bounds based on Inductive Conformal Prediction (ICP). We train a Triplet Network architecture to learn representations of the input data that can be used to estimate the similarity between test examples and examples in the training data set. Then, these representations are used to estimate the confidence of set predictions from a classifier that is based on the neural network architecture used in the triplet. The approach is evaluated using a robotic navigation benchmark and the results show that we can computed trusted confidence bounds efficiently in real-time.
Learning-enabled components (LECs) are widely used in cyber-physical systems (CPS) since they can handle the uncertainty and variability of the environment and increase the level of autonomy. However, it has been shown that LECs such as deep neural networks (DNN) are not robust and adversarial examples can cause the model to make a false prediction. The paper considers the problem of efficiently detecting adversarial examples in LECs used for regression in CPS. The proposed approach is based on inductive conformal prediction and uses a regression model based on variational autoencoder. The architecture allows to take into consideration both the input and the neural network prediction for detecting adversarial, and more generally, out-of-distribution examples. We demonstrate the method using an advanced emergency braking system implemented in an open source simulator for self-driving cars where a DNN is used to estimate the distance to an obstacle. The simulation results show that the method can effectively detect adversarial examples with a short detection delay.
Cyber-Physical Systems (CPS) have been increasingly subject to cyber-attacks including code injection attacks. Zero day attacks further exasperate the threat landscape by requiring a shift to defense in depth approaches. With the tightly coupled nature of cyber components with the physical domain, these attacks have the potential to cause significant damage if safety-critical applications such as automobiles are compromised. Moving target defense techniques such as instruction set randomization (ISR) have been commonly proposed to address these types of attacks. However, under current implementations an attack can result in system crashing which is unacceptable in CPS. As such, CPS necessitate proper control reconfiguration mechanisms to prevent a loss of availability in system operation. This paper addresses the problem of maintaining system and security properties of a CPS under attack by integrating ISR, detection, and recovery capabilities that ensure safe, reliable, and predictable system operation. Specifically, we consider the problem of detecting code injection attacks and reconfiguring the controller in real-time. The developed framework is demonstrated with an autonomous vehicle case study.
Smart water networks can provide great benefits to our society in terms of efficiency and sustainability. However, smart capabilities and connectivity also expose these systems to a wide range of cyber attacks, which enable cyber-terrorists and hostile nation states to mount cyber-physical attacks. Cyber-physical attacks against critical infrastructure, such as water treatment and distribution systems, pose a serious threat to public safety and health. Consequently, it is imperative that we improve the resilience of smart water networks. We consider three approaches for improving resilience: redundancy, diversity, and hardening. Even though each one of these "canonical" approaches has been throughly studied in prior work, a unified theory on how to combine them in the most efficient way has not yet been established. In this paper, we address this problem by studying the synergy of these approaches in the context of protecting smart water networks from cyber-physical contamination attacks.
In this paper, we study the sensor placement problem in urban water networks that maximizes the localization of pipe failures given that some sensors give incorrect outputs. False output of a sensor might be the result of degradation in sensor's hardware, software fault, or might be due to a cyber attack on the sensor. Incorrect outputs from such sensors can have any possible values which could lead to an inaccurate localization of a failure event. We formulate the optimal sensor placement problem with erroneous sensors as a set multicover problem, which is NP-hard, and then discuss a polynomial time heuristic to obtain efficient solutions. In this direction, we first examine the physical model of the disturbance propagating in the network as a result of a failure event, and outline the multi-level sensing model that captures several event features. Second, using a combinatorial approach, we solve the problem of sensor placement that maximizes the localization of pipe failures by selecting m sensors out of which at most e give incorrect outputs. We propose various localization performance metrics, and numerically evaluate our approach on a benchmark and a real water distribution network. Finally, using computational experiments, we study relationships between design parameters such as the total number of sensors, the number of sensors with errors, and extracted signal features.
One step involved in the security engineering process is threat modeling. Threat modeling involves understanding the complexity of the system and identifying all of the possible threats, regardless of whether or not they can be exploited. Proper identification of threats and appropriate selection of countermeasures reduces the ability of attackers to misuse the system. This paper presents a quantitative, integrated threat modeling approach that merges software and attack centric threat modeling techniques. The threat model is composed of a system model representing the physical and network infrastructure layout, as well as a component model illustrating component specific threats. Component attack trees allow for modeling specific component contained attack vectors, while system attack graphs illustrate multi-component, multi-step attack vectors across the system. The Common Vulnerability Scoring System (CVSS) is leveraged to provide a standardized method of quantifying the low level vulnerabilities in the attack trees. As a case study, a railway communication network is used, and the respective results using a threat modeling software tool are presented.
In recent years, we have seen a number of successful attacks against high-profile targets, some of which have even caused severe physical damage. These examples have shown us that resourceful and determined attackers can penetrate virtually any system, even those that are secured by the "air-gap." Consequently, in order to minimize the impact of stealthy attacks, defenders have to focus not only on strengthening the first lines of defense but also on deploying effective intrusion-detection systems. Intrusion-detection systems can play a key role in protecting sensitive computer systems since they give defenders a chance to detect and mitigate attacks before they could cause substantial losses. However, an over-sensitive intrusion-detection system, which produces a large number of false alarms, imposes prohibitively high operational costs on a defender since alarms need to be manually investigated. Thus, defenders have to strike the right balance between maximizing security and minimizing costs. Optimizing the sensitivity of intrusion detection systems is especially challenging in the case when multiple inter-dependent computer systems have to be defended against a strategic attacker, who can target computer systems in order to maximize losses and minimize the probability of detection. We model this scenario as an attacker-defender security game and study the problem of finding optimal intrusion detection thresholds.
In-depth consideration and evaluation of security and resilience is necessary for developing the scientific foundations and technology of Cyber-Physical Systems (CPS). In this demonstration, we present SURE [1], a CPS experimentation and evaluation testbed for security and resilience focusing on transportation networks. The testbed includes (1) a heterogeneous modeling and simulation integration platform, (2) a Web-based tool for modeling CPS in adversarial environments, and (3) a framework for evaluating resilience using attacker-defender games. Users such as CPS designers and operators can interact with the testbed to evaluate monitoring and control schemes that include sensor placement and traffic signal configuration.
In order to be resilient to attacks, a cyber-physical system (CPS) must be able to detect attacks before they can cause significant damage. To achieve this, \emph{intrusion detection systems} (IDS) may be deployed, which can detect attacks and alert human operators, who can then intervene. However, the resource-constrained nature of many CPS poses a challenge, since reliable IDS can be computationally expensive. Consequently, computational nodes may not be able to perform intrusion detection continuously, which means that we have to devise a schedule for performing intrusion detection. While a uniformly random schedule may be optimal in a purely cyber system, an optimal schedule for protecting CPS must also take into account the physical properties of the system, since the set of adversarial actions and their consequences depend on the physical systems. Here, in the context of water distribution networks, we study IDS scheduling problems in two settings and under the constraints on the available battery supplies. In the first problem, the objective is to design, for a given duration of time $T$, scheduling schemes for IDS so that the probability of detecting an attack is maximized within that duration. We propose efficient heuristic algorithms for this general problem and evaluate them on various networks. In the second problem, our objective is to design scheduling schemes for IDS so that the overall lifetime of the network is maximized while ensuring that an intruder attack is always detected. Various strategies to deal with this problem are presented and evaluated for various networks.
The objective of this work is to develop an efficient and practical sensor placement method for the failure detection and localization in water networks. We formulate the problem as the minimum test cover problem (MTC) with the objective of selecting the minimum number of sensors required to uniquely identify and localize pipe failure events. First, we summarize a single-level sensing model and discuss an efficient fast greedy approach for solving the MTC problem. Simulation results on benchmark test networks demonstrate the efficacy of the fast greedy algorithm. Second, we develop a multi-level sensing model that captures additional physical features of the disturbance event, such as the time lapsed between the occurrence of disturbance and its detection by the sensor. Our sensor placement approach using MTC extends to the multi-level sensing model and an improved identification performance is obtained via reduced number of sensors (in comparison to single-level sensing model). In particular, we investigate the bi-level sensing model to illustrate the efficacy of employing multi-level sensors for the identification of failure events. Finally, we suggest extensions of our approach for the deployment of heterogeneous sensors in water networks by exploring the trade-off between cost and performance (measured in terms of the identification score of pipe/link failures).
Due to their low deployment costs, wireless sensor networks (WSN) may act as a key enabling technology for a variety of spatially-distributed cyber-physical system (CPS) applications, ranging from intelligent traffic control to smart grids. However, besides providing tremendous benefits in terms of deployment costs, they also open up new possibilities for malicious attackers, who aim to cause financial losses or physical damage. Since perfectly securing these spatially-distributed systems is either impossible or financially unattainable, we need to design them to be resilient to attacks: even if some parts of the system are compromised or unavailable due to the actions of an attacker, the system as a whole must continue to operate with minimal losses. In a CPS, control decisions affecting the physical process depend on the observed data from the sensor network. Any malicious activity in the sensor network can therefore severely impact the physical process, and consequently the overall CPS operations. These factors necessitate a deeper probe into the domain of resilient WSN for CPS. In this chapter, we provide an overview of various dimensions in this field, including objectives of WSN in CPS, attack scenarios and vulnerabilities, notion of attack-resilience in WSN for CPS, and solution approaches towards attaining resilience. We also highlight major challenges, recent developments, and future directions in this area.
To penetrate sensitive computer networks, attackers can use spear phishing to sidestep technical security mechanisms by exploiting the privileges of careless users. In order to maximize their success probability, attackers have to target the users that constitute the weakest links of the system. The optimal selection of these target users takes into account both the damage that can be caused by a user and the probability of a malicious e-mail being delivered to and opened by a user. Since attackers select their targets in a strategic way, the optimal mitigation of these attacks requires the defender to also personalize the e-mail filters by taking into account the users' properties.
In this paper, we assume that a learned classifier is given and propose strategic per-user filtering thresholds for mitigating spear-phishing attacks. We formulate the problem of filtering targeted and non-targeted malicious e-mails as a Stackelberg security game. We characterize the optimal filtering strategies and show how to compute them in practice. Finally, we evaluate our results using two real-world datasets and demonstrate that the proposed thresholds lead to lower losses than nonstrategic thresholds.
Assuring communication integrity is a central problem in security. However, overhead costs associated with cryptographic primitives used towards this end introduce significant practical implementation challenges for resource-bounded systems, such as cyber-physical systems. For example, many control systems are built on legacy components which are computationally limited but have strict timing constraints. If integrity protection is a binary decision, it may simply be infeasible to introduce into such systems; without it, however, an adversary can forge malicious messages, which can cause signicant physical or financial harm. We propose a formal game-theoretic framework for optimal stochastic message authentication, providing provable integrity guarantees for resource-bounded systems based on an existing MAC scheme. We use our framework to investigate attacker deterrence, as well as optimal design of stochastic message authentication schemes when deterrence is impossible. Finally, we provide experimental results on the computational performance of our framework in practice.