Martín Ochoa

Ann Yi Wong, Eyasu Getahun Chekole, Martin Ochoa et al.

Traditionally, applications that are used in large and small enterprises were deployed on "bare metal" servers installed with operating systems. Recently, the use of multiple virtual machines (VMs) on the same physical server was adopted due to cost reduction and flexibility. Nowadays, containers have become popular for application deployment due to smaller footprints than the VMs, their ability to start and stop more quickly, and their capability to pack the application binaries and their dependencies in standalone units for seamless portability. A typical container ecosystem includes a code repository (e.g., GitHub) where the container images are built from the codes and libraries and then pushed to the image registry (e.g., Docker Hub) for subsequent deployment as application containers. However, the pervasive use of containers also leads to a wide-range of security breaches, such as stealing credentials and sensitive data from image registry and code repository, carrying out DoS attacks, and gaining root access to the underlying host. In this paper, we first perform threat modeling on the containers ecosystem using the popular threat modeling framework, called STRIDE. Using STRIDE, we identify the vulnerabilities in each system component, and investigate potential security threats and their consequences. Then, we conduct a comprehensive survey on the existing countermeasures designed against the identified threats and vulnerabilities. In particular, we assess the strengths and weaknesses of the existing mitigation strategies designed against such threats. We believe that this work will help researchers and practitioners to gain a deeper understanding of the threat landscape in containers and the state-of-the-art countermeasures. We also discuss open research problems and future research directions in containers security, which may ignite further research to be done in this area.

Jakub Breier, Xiaolu Hou, Martín Ochoa et al.

Neural network implementations are known to be vulnerable to physical attack vectors such as fault injection attacks. As of now, these attacks were only utilized during the inference phase with the intention to cause a misclassification. In this work, we explore a novel attack paradigm by injecting faults during the training phase of a neural network in a way that the resulting network can be attacked during deployment without the necessity of further faulting. In particular, we discuss attacks against ReLU activation functions that make it possible to generate a family of malicious inputs, which are called fooling inputs, to be used at inference time to induce controlled misclassifications. Such malicious inputs are obtained by mathematically solving a system of linear equations that would cause a particular behaviour on the attacked activation functions, similar to the one induced in training through faulting. We call such attacks fooling backdoors as the fault attacks at the training phase inject backdoors into the network that allow an attacker to produce fooling inputs. We evaluate our approach against multi-layer perceptron networks and convolutional networks on a popular image classification task obtaining high attack success rates (from 60% to 100%) and high classification confidence when as little as 25 neurons are attacked while preserving high accuracy on the originally intended classification task.

Chuadhry Mujeeb Ahmed, Martin Ochoa, Jianying Zhou et al.

Programmable Logic Controllers (PLCs) are a core component of an Industrial Control System (ICS). However, if a PLC is compromised or the commands sent across a network from the PLCs are spoofed, consequences could be catastrophic. In this work, a novel technique to authenticate PLCs is proposed that aims at raising the bar against powerful attackers while being compatible with real-time systems. The proposed technique captures timing information for each controller in a non-invasive manner. It is argued that Scan Cycle is a unique feature of a PLC that can be approximated passively by observing network traffic. An attacker that spoofs commands issued by the PLCs would deviate from such fingerprints. To detect replay attacks a PLC Watermarking technique is proposed. PLC Watermarking models the relationship between the scan cycle and the control logic by modeling the input/output as a function of request/response messages of a PLC. The proposed technique is validated on an operational water treatment plant (SWaT) and smart grid (EPIC) testbed. Results from experiments indicate that PLCs can be distinguished based on their scan cycle timing characteristics.

Eyasu Getahun Chekole, Martin Ochoa, Sudipta Chattopadhyay

Cyber-Physical Systems (CPS) are being widely adopted in critical infrastructures, such as smart grids, nuclear plants, water systems, transportation systems, manufacturing and healthcare services, among others. However, the increasing prevalence of cyberattacks targeting them raises a growing security concern in the domain. In particular, memory-safety attacks, that exploit memory-safety vulnerabilities, constitute a major attack vector against real-time control devices in CPS. Traditional IT countermeasures against such attacks have limitations when applied to the CPS context: they typically incur in high runtime overheads; which conflicts with real-time constraints in CPS and they often abort the program when an attack is detected, thus harming availability of the system, which in turn can potentially result in damage to the physical world. In this work, we propose to enforce a full-stack memory-safety (covering user-space and kernel-space attack surfaces) based on secure compiling of PLCs to detect memory-safety attacks in CPS. Furthermore, to ensure availability, we enforce a resilient mitigation technique that bypasses illegal memory access instructions at runtime by dynamically instrumenting low-level code. We empirically measure the computational overhead caused by our approach on two experimental settings based on real CPS. The experimental results show that our approach effectively and efficiently detects and mitigates memory-safety attacks in realistic CPS.

Eyasu Getahun Chekole, Unnikrishnan Cheramangalath, Sudipta Chattopadhyay et al.

Memory-safety attacks have been one of the most critical threats against computing systems. Although a wide-range of defense techniques have been developed against these attacks, the existing mitigation strategies have several limitations. In particular, most of the existing mitigation approaches are based on aborting or restarting the victim program when a memory-safety attack is detected, thus making the system unavailable. This might not be acceptable in systems with stringent timing constraints, such as cyber-physical systems (CPS), since the system unavailability leaves the control system in an unsafe state. To address this problem, we propose CIMA -- a resilient and light-weight mitigation technique that prevents invalid memory accesses at runtime. CIMA manipulates the compiler-generated control flow graph to automatically detect and bypass unsafe memory accesses at runtime, thereby mitigating memory-safety attacks along the process. An appealing feature of CIMA is that it also ensures system availability and resilience of the CPS even under the presence of memory-safety attacks. To this end, we design our experimental setup based on a realistic Secure Water Treatment (SWaT) and Secure Urban Transportation System (SecUTS) testbeds and evaluate the effectiveness and the efficiency of our approach. The experimental results reveal that CIMA handles memory-safety attacks effectively with low overhead. Moreover, it meets the real-time constraints and physical-state resiliency of the CPS under test.

Ivan Homoliak, Flavio Toffalini, Juan Guarnizo et al.

Insider threats are one of today's most challenging cybersecurity issues that are not well addressed by commonly employed security solutions. Despite several scientific works published in this domain, we argue that the field can benefit from the proposed structural taxonomy and novel categorization of research that contribute to the organization and disambiguation of insider threat incidents and the defense solutions used against them. The objective of our categorization is to systematize knowledge in insider threat research, while leveraging existing grounded theory method for rigorous literature review. The proposed categorization depicts the workflow among particular categories that include: 1) Incidents and datasets, 2) Analysis of attackers, 3) Simulations, and 4) Defense solutions. Special attention is paid to the definitions and taxonomies of the insider threat; we present a structural taxonomy of insider threat incidents, which is based on existing taxonomies and the 5W1H questions of the information gathering problem. Our survey will enhance researchers' efforts in the domain of insider threat, because it provides: a) a novel structural taxonomy that contributes to orthogonal classification of incidents and defining the scope of defense solutions employed against them, b) an updated overview on publicly available datasets that can be used to test new detection solutions against other works, c) references of existing case studies and frameworks modeling insiders' behaviors for the purpose of reviewing defense solutions or extending their coverage, and d) a discussion of existing trends and further research directions that can be used for reasoning in the insider threat domain.

Chuadhry Mujeeb Ahmed, Aditya Mathur, Martin Ochoa

In recent years fingerprinting of various physical and logical devices has been proposed with the goal of uniquely identifying users or devices of mainstream IT systems such as PCs, Laptops and smart phones. On the other hand, the application of such techniques in Cyber-Physical Systems (CPS) is less explored due to various reasons, such as difficulty of direct access to critical systems and the cost involved in faithfully reproducing realistic scenarios. In this work we evaluate the feasibility of using fingerprinting techniques in the context of realistic Industrial Control Systems related to water treatment and distribution. Based on experiments conducted with 44 sensors of six different types, it is shown that noise patterns due to microscopic imperfections in hardware manufacturing can be used to uniquely identify sensors in a CPS with up to 97% accuracy. The proposed technique can be used in to detect physical attacks, such as the replacement of legitimate sensors by faulty or manipulated sensors. We also show that, unexpectedly, sensor fingerprinting can effectively detect advanced physical attacks such as analog sensor spoofing due to variations in received energy at the transducer of an active sensor. Also, it can be leveraged to construct a novel challenge-response protocol that exposes cyber-attacks.

Yair Meidan, Michael Bohadana, Asaf Shabtai et al.

Security experts have demonstrated numerous risks imposed by Internet of Things (IoT) devices on organizations. Due to the widespread adoption of such devices, their diversity, standardization obstacles, and inherent mobility, organizations require an intelligent mechanism capable of automatically detecting suspicious IoT devices connected to their networks. In particular, devices not included in a white list of trustworthy IoT device types (allowed to be used within the organizational premises) should be detected. In this research, Random Forest, a supervised machine learning algorithm, was applied to features extracted from network traffic data with the aim of accurately identifying IoT device types from the white list. To train and evaluate multi-class classifiers, we collected and manually labeled network traffic data from 17 distinct IoT devices, representing nine types of IoT devices. Based on the classification of 20 consecutive sessions and the use of majority rule, IoT device types that are not on the white list were correctly detected as unknown in 96% of test cases (on average), and white listed device types were correctly classified by their actual types in 99% of cases. Some IoT device types were identified quicker than others (e.g., sockets and thermostats were successfully detected within five TCP sessions of connecting to the network). Perfect detection of unauthorized IoT device types was achieved upon analyzing 110 consecutive sessions; perfect classification of white listed types required 346 consecutive sessions, 110 of which resulted in 99.49% accuracy. Further experiments demonstrated the successful applicability of classifiers trained in one location and tested on another. In addition, a discussion is provided regarding the resilience of our machine learning-based IoT white listing method to adversarial attacks.

Eric Rothstein Morris, Carlos G. Murguia, Martín Ochoa

In a software system it is possible to quantify the amount of information that is leaked or corrupted by analysing the flows of information present in the source code. In a cyber-physical system, information flows are not only present at the digital level, but also at a physical level, and to and fro the two levels. In this work, we provide a methodology to formally analyse a Cyber-Physical System composite model (combining physics and control) using an information flow-theoretic approach. We use this approach to quantify the level of vulnerability of a system with respect to attackers with different capabilities. We illustrate our approach by means of a water distribution case study.

Daniele Antonioli, Hamid Reza Ghaeini, Sridhar Adepu et al.

In this work, we consider challenges relating to security for Industrial Control Systems (ICS) in the context of ICS security education and research targeted both to academia and industry. We propose to address those challenges through gamified attack training and countermeasure evaluation. We tested our proposed ICS security gamification idea in the context of the (to the best of our knowledge) first Capture-The-Flag (CTF) event targeted to ICS security called SWaT Security Showdown (S3). Six teams acted as attackers in a security competition leveraging an ICS testbed, with several academic defense systems attempting to detect the ongoing attacks. The event was conducted in two phases. The online phase (a jeopardy-style CTF) served as a training session. The live phase was structured as an attack-defense CTF. We acted as judges and we assigned points to the attacker teams according to a scoring system that we developed internally based on multiple factors, including realistic attacker models. We conclude the paper with an evaluation and discussion of the S3, including statistics derived from the data collected in each phase of S3.

Martín Ochoa, Sebastian Banescu, Cynthia Disenfeld et al.

Despite numerous countermeasures proposed by practitioners and researchers, remote control-flow alteration of programs with memory-safety vulnerabilities continues to be a realistic threat. Guaranteeing that complex software is completely free of memory-safety vulnerabilities is extremely expensive. Probabilistic countermeasures that depend on random secret keys are interesting, because they are an inexpensive way to raise the bar for attackers who aim to exploit memory-safety vulnerabilities. Moreover, some countermeasures even support legacy systems. However, it is unclear how to quantify and compare the effectiveness of different probabilistic countermeasures or combinations of such countermeasures. In this paper we propose a methodology to rigorously derive security bounds for probabilistic countermeasures. We argue that by representing security notions in this setting as events in probabilistic games, similarly as done with cryptographic security definitions, concrete and asymptotic guarantees can be obtained against realistic attackers. These guarantees shed light on the effectiveness of single countermeasures and their composition and allow practitioners to more precisely gauge the risk of an attack.

Juan Guarnizo, Amit Tambe, Suman Sankar Bhunia et al.

In recent years, the emerging Internet-of-Things (IoT) has led to rising concerns about the security of networked embedded devices. In this work, we focus on the adaptation of Honeypots for improving the security of IoTs. Low-interaction honeypots are used so far in the context of IoT. Such honeypots are limited and easily detectable, and thus, there is a need to find ways how to develop high-interaction, reliable, IoT honeypots that will attract skilled attackers. In this work, we propose the SIPHON architecture - a Scalable high-Interaction Honeypot platform for IoT devices. Our architecture leverages IoT devices that are physically at one location and are connected to the Internet through so-called wormholes distributed around the world. The resulting architecture allows exposing few physical devices over a large number of geographically distributed IP addresses. We demonstrate the proposed architecture in a large scale experiment with 39 wormhole instances in 16 cities in 9 countries. Based on this setup, six physical IP cameras, one NVR and one IP printer are presented as 85 real IoT devices on the Internet, attracting a daily traffic of 700MB for a period of two months. A preliminary analysis of the collected traffic indicates that devices in some cities attracted significantly more traffic than others (ranging from 600 000 incoming TCP connections for the most popular destination to less than 50000 for the least popular). We recorded over 400 brute-force login attempts to the web-interface of our devices using a total of 1826 distinct credentials, from which 11 attempts were successful. Moreover, we noted login attempts to Telnet and SSH ports some of which used credentials found in the recently disclosed Mirai malware.

Vijay Ganesh, Sebastian Banescu, Martín Ochoa

In this paper, we introduce a formal notion of partial compliance, called Attack-resistance, of a computer program running together with a defense mechanism w.r.t a non-exploitability specification. In our setting, a program may contain exploitable vulnerabilities, such as buffer overflows, but appropriate defense mechanisms built into the program or the operating system render such vulnerabilities hard to exploit by certain attackers, usually relying on the strength of the randomness of a probabilistic transformation of the environment or the program and some knowledge on the attacker's goals and attack strategy. We are motivated by the reality that most large-scale programs have vulnerabilities despite our best efforts to get rid of them. Security researchers have responded to this state of affairs by coming up with ingenious defense mechanisms such as address space layout randomization (ASLR) or instruction set randomization (ISR) that provide some protection against exploitation. However, implementations of such mechanism have been often shown to be insecure, even against the attacks they were designed to prevent. By formalizing this notion of attack-resistance we pave the way towards addressing the questions: "How do we formally analyze defense mechanisms? Is there a mathematical way of distinguishing effective defense mechanisms from ineffective ones? Can we quantify and show that these defense mechanisms provide formal security guarantees, albeit partial, even in the presence of exploitable vulnerabilities?". To illustrate our approach we discuss under which circumstances ISR implementations comply with the Attack-resistance definition.

Sebastian Banescu, Tobias Wüchner, Marius Guggenmos et al.

Program obfuscation is increasingly popular among malware creators. Objectively comparing different malware detection approaches with respect to their resilience against obfuscation is challenging. To the best of our knowledge, there is no common empirical framework for evaluating the resilience of malware detection approaches w.r.t. behavior obfuscation. We propose and implement such a framework that obfuscates the observable behavior of malware binaries. To assess the framework's utility, we use it to obfuscate known malware binaries and then investigate the impact on detection effectiveness of different $n$-gram based detection approaches. We find that the obfuscation transformations employed by our framework significantly affect the precision of such detection approaches. Several $n$-gram-based approaches can hence be concluded not to be resilient against this simple kind of obfuscation.

Tobias Wüchner, Martín Ochoa, Alexander Pretschner

We present a novel malware detection approach based on metrics over quantitative data flow graphs. Quantitative data flow graphs (QDFGs) model process behavior by interpreting issued system calls as aggregations of quantifiable data flows.Due to the high abstraction level we consider QDFG metric based detection more robust against typical behavior obfuscation like bogus call injection or call reordering than other common behavioral models that base on raw system calls. We support this claim with experiments on obfuscated malware logs and demonstrate the superior obfuscation robustness in comparison to detection using n-grams. Our evaluations on a large and diverse data set consisting of about 7000 malware and 500 goodware samples show an average detection rate of 98.01% and a false positive rate of 0.48%. Moreover, we show that our approach is able to detect new malware (i.e. samples from malware families not included in the training set) and that the consideration of quantities in itself significantly improves detection precision.