Biblio

Identity-Based Encryption (IBE) was introduced as an elegant concept for secure data exchange due to its simplified key management by specifically addressing the asymmetric key distribution problems in multi-user scenarios. In the context of ad-hoc network connections that are of particular importance in the emerging Internet of Things, the simple key discovery procedures as provided by IBE are very beneficial in many situations. In this work we demonstrate for the first time that IBE has become practical even for a range of embedded devices that are populated with low-cost ARM Cortex-M microcontrollers or reconfigurable hardware components. More precisely, we adopt the IBE scheme proposed by Ducas et al. at ASIACRYPT 2014 based on the RLWE problem for which we provide implementation results for two security levels on the aforementioned embedded platforms. We give evidence that the implementations of the basic scheme are efficient, as for a security level of 80 bits it requires 103 ms and 36 ms for encryption and decryption, respectively, on the smallest ARM Cortex-M0 microcontroller.

A number of online services nowadays rely upon machine learning to extract valuable information from data collected in the wild. This exposes learning algorithms to the threat of data poisoning, i.e., a coordinate attack in which a fraction of the training data is controlled by the attacker and manipulated to subvert the learning process. To date, these attacks have been devised only against a limited class of binary learning algorithms, due to the inherent complexity of the gradient-based procedure used to optimize the poisoning points (a.k.a. adversarial training examples). In this work, we first extend the definition of poisoning attacks to multiclass problems. We then propose a novel poisoning algorithm based on the idea of back-gradient optimization, i.e., to compute the gradient of interest through automatic differentiation, while also reversing the learning procedure to drastically reduce the attack complexity. Compared to current poisoning strategies, our approach is able to target a wider class of learning algorithms, trained with gradient-based procedures, including neural networks and deep learning architectures. We empirically evaluate its effectiveness on several application examples, including spam filtering, malware detection, and handwritten digit recognition. We finally show that, similarly to adversarial test examples, adversarial training examples can also be transferred across different learning algorithms.

With the rapid development of sophisticated attack techniques, individual security systems that base all of their decisions and actions of attack prevention and response on their own observations and knowledge become incompetent. To cope with this problem, collaborative security in which a set of security entities are coordinated to perform specific security actions is proposed in literature. In collaborative security schemes, multiple entities collaborate with each other by sharing threat evidence or analytics to make more effective decisions. Nevertheless, the anticipated information exchange raises privacy concerns, especially for those privacy-sensitive entities. In order to obtain a quantitative understanding of the fundamental tradeoff between the effectiveness of collaboration and the entities' privacy, a repeated two-layer single-leader multi-follower game is proposed in this work. Based on our game-theoretic analysis, the expected behaviors of both the attacker and the security entities are derived and the utility-privacy tradeoff curve is obtained. In addition, the existence of Nash equilibrium (NE) for the collaborative entities is proven, and an asynchronous dynamic update algorithm is proposed to compute the optimal collaboration strategies of the entities. Furthermore, the existence of Byzantine entities is considered and its influence is investigated. Finally, simulation results are presented to validate the analysis.

SDN networks rely mainly on a set of software defined modules, running on generic hardware platforms, and managed by a central SDN controller. The tight coupling and lack of isolation between the controller and the underlying host limit the controller resilience against host-based attacks and failures. That controller is a single point of failure and a target for attackers. ``Linux-containers'' is a successful thin virtualization technique that enables encapsulated, host-isolated execution-environments for running applications. In this paper we present PAFR, a controller sandboxing mechanism based on Linux-containers. PAFR enables controller/host isolation, plug-and-play operation, failure-and-attack-resilient execution, and fast recovery. PAFR employs and manages live remote checkpointing and migration between different hosts to evade failures and attacks. Experiments and simulations show that the frequent employment of PAFR's live-migration minimizes the chance of successful attack/failure with limited to no impact on network performance.

Nowadays the adoption of IoT solutions is gaining high momentum in several fields, including energy, home and environment monitoring, transportation, and manufacturing. However, cybersecurity attacks to low-cost end-user devices can severely undermine the expected deployment of IoT solutions in a broad range of scenarios. To face these challenges, emerging software-based networking features can introduce new security enablers, providing further scalability and flexibility required to cope with massive IoT. In this paper, we present a novel framework aiming to exploit SDN/NFV-based security features and devise new efficient integration with existing IoT security approaches. The potential benefits of the proposed framework is validated in two case studies. Finally, a feasibility study is presented, accounting for potential interactions with open-source SDN/NFV projects and relevant standardization activities.

A long time ago Industrial Control Systems were in a safe place due to the use of proprietary technology and physical isolation. This situation has changed dramatically and the systems are nowadays often prone to severe attacks executed from remote locations. In many cases, intrusions remain undetected for a long time and this allows the adversary to meticulously prepare an attack and maximize its destructiveness. The ability to detect an attack in its early stages thus has a high potential to significantly reduce its impact. To this end, we propose a holistic, multi-layered, security monitoring and mitigation framework spanning the physical- and cyber domain. The comprehensiveness of the approach demands for scalability measures built-in by design. In this paper we present how scalability is addressed by an architecture that enforces geographically decentralized data reduction approaches that can be dynamically adjusted to the currently perceived context. A specific focus is put on a robust and resilient solution to orchestrate dynamic configuration updates. Experimental results based on a prototype implementation show the feasibility of the approach.

Enhancing trust among service providers and end-users with respect to data protection is an urgent matter in the growing information society. In response, CREDENTIAL proposes an innovative cloud-based service for storing, managing, and sharing of digital identity information and other highly critical personal data with a demonstrably higher level of security than other current solutions. CREDENTIAL enables end-to-end confidentiality and authenticity as well as improved privacy in cloud-based identity management and data sharing scenarios. In this paper, besides clarifying the vision and use cases, we focus on the adoption of CREDENTIAL. Firstly, for adoption by providers, we elaborate on the functionality of CREDENTIAL, the services implementing these functions, and the physical architecture needed to deploy such services. Secondly, we investigate factors from related research that could be used to facilitate CREDENTIAL's adoption and list key benefits as convincing arguments.

The importance of Networked Control Systems (NCS) is steadily increasing due to recent trends such as smart factories. Correct functionality of such NCS needs to be protected as malfunctioning systems could have severe consequences for the controlled process or even threaten human lives. However, with the increase in NCS, also attacks targeting these systems are becoming more frequent. To mitigate attacks that utilize captured sensor data in an NCS, transferred data needs to be protected. While using well-known methods such as Transport Layer Security (TLS) might be suitable to protect the data, resource constraint devices such as sensors often are not powerful enough to perform the necessary cryptographic operations. Also, as we will show in this paper, applying simple encryption in an NCS may enable easy Denial-of-Service (DoS) attacks by attacking single bits of the encrypted data. Therefore, in this paper, we present a hardware-based approach that enables sensors to perform the necessary encryption while being robust against (injected) bit failures.

In this paper, we describe an efficient methodology to guide investigators during network forensic analysis. To this end, we introduce the concept of core attack graph, a compact representation of the main routes an attacker can take towards specific network targets. Such compactness allows forensic investigators to focus their efforts on critical nodes that are more likely to be part of attack paths, thus reducing the overall number of nodes (devices, network privileges) that need to be examined. Nevertheless, core graphs also allow investigators to hierarchically explore the graph in order to retrieve different levels of summarised information. We have evaluated our approach over different network topologies varying parameters such as network size, density, and forensic evaluation threshold. Our results demonstrate that we can achieve the same level of accuracy provided by standard logical attack graphs while significantly reducing the exploration rate of the network.

Critical business applications in domains ranging from technical support to healthcare increasingly rely on large-scale, automatically constructed knowledge graphs. These applications use the results of complex queries over knowledge graphs in order to help users in taking crucial decisions such as which drug to administer, or whether certain actions are compliant with all the regulatory requirements and so on. However, these knowledge graphs constantly evolve, and the newer versions may adversely impact the results of queries that the previously taken business decisions were based on. We propose a framework based on provenance polynomials to track the impact of knowledge graph changes on arbitrary SPARQL query results. Focusing on the deletion of facts, we show how to efficiently determine the queries impacted by the change, develop ways to incrementally maintain these polynomials, and present an efficient implementation on top of RDF graph databases. Our experimental evaluation over large-scale RDF/SPARQL benchmarks show the effectiveness of our proposal.

Intrusion Detection Systems (IDSs) are crucial security mechanisms widely deployed for critical network protection. However, conventional IDSs become incompetent due to the rapid growth in network size and the sophistication of large scale attacks. To mitigate this problem, Collaborative IDSs (CIDSs) have been proposed in literature. In CIDSs, a number of IDSs exchange their intrusion alerts and other relevant data so as to achieve better intrusion detection performance. Nevertheless, the required information exchange may result in privacy leakage, especially when these IDSs belong to different self-interested organizations. In order to obtain a quantitative understanding of the fundamental tradeoff between the intrusion detection accuracy and the organizations' privacy, a repeated two-layer single-leader multi-follower game is proposed in this work. Based on our game-theoretic analysis, we are able to derive the expected behaviors of both the attacker and the IDSs and obtain the utility-privacy tradeoff curve. In addition, the existence of Nash equilibrium (NE) is proved and an asynchronous dynamic update algorithm is proposed to compute the optimal collaboration strategies of IDSs. Finally, simulation results are shown to validate the analysis.

Intrusion Detection Systems (IDSs) are crucial security mechanisms widely deployed for critical network protection. However, conventional IDSs become incompetent due to the rapid growth in network size and the sophistication of large scale attacks. To mitigate this problem, Collaborative IDSs (CIDSs) have been proposed in literature. In CIDSs, a number of IDSs exchange their intrusion alerts and other relevant data so as to achieve better intrusion detection performance. Nevertheless, the required information exchange may result in privacy leakage, especially when these IDSs belong to different self-interested organizations. In order to obtain a quantitative understanding of the fundamental tradeoff between the intrusion detection accuracy and the organizations' privacy, a repeated two-layer single-leader multi-follower game is proposed in this work. Based on our game-theoretic analysis, we are able to derive the expected behaviors of both the attacker and the IDSs and obtain the utility-privacy tradeoff curve. In addition, the existence of Nash equilibrium (NE) is proved and an asynchronous dynamic update algorithm is proposed to compute the optimal collaboration strategies of IDSs. Finally, simulation results are shown to validate the analysis.

This paper develops methods to efficiently compute the set of disturbances on a power network that do not tip the frequency of each bus and the power flow in each transmission line beyond their respective bounds. For a linearized AC power network model, we propose a sampling method to provide superset and subset approximations with a desired accuracy of the set of feasible disturbances. We also introduce an error metric to measure the approximation gap and design an algorithm that is able to reduce its value without impacting the complexity of the resulting set approximations. Simulations on the IEEE 118-bus power network illustrate our results.