Iakovos Gurulian

Secure and Trusted Execution: Past, Present, and Future - A Critical Review in the Context of the Internet of Things and Cyber-Physical Systems

Notions like security, trust, and privacy are crucial in the digital environment and in the future, with the advent of technologies like the Internet of Things (IoT) and Cyber-Physical Systems (CPS), their importance is only going to increase. Trust has different definitions, some situations rely on real-world relationships between entities while others depend on robust technologies to gain trust after deployment. In this paper we focus on these robust technologies, their evolution in past decades and their scope in the near future. The evolution of robust trust technologies has involved diverse approaches, as a consequence trust is defined, understood and ascertained differently across heterogeneous domains and technologies. In this paper we look at digital trust technologies from the point of view of security and examine how they are making secure computing an attainable reality. The paper also revisits and analyses the Trusted Platform Module (TPM), Secure Elements (SE), Hypervisors and Virtualisation, Intel TXT, Trusted Execution Environments (TEE) like GlobalPlatform TEE, Intel SGX, along with Host Card Emulation, and Encrypted Execution Environment (E3). In our analysis we focus on these technologies and their application to the emerging domains of the IoT and CPS.

You can’t touch this: Consumer-centric android application repackaging detection

Abstract Application repackaging is a widely used method for malware distribution, revenue stealing and piracy. Repackaged applications are modified versions of original applications, that can potentially target large audiences based on the original application’s popularity. In this paper, we propose an approach for detecting repackaged applications. Our approach takes advantage of the attacker’s reluctance to significantly alter the elements that characterise an application without notably impacting the application’s distribution. These elements include the application’s name and icon. The detection is initiated from the client side, prior to an application’s installation, making it application store agnostic. Our experimental results show that detection based on our algorithm is effective and efficient.

Preventing relay attacks in mobile transactions using infrared light

Near Field Technology (NFC) enables a smartphone to emulate a smart card, enabling it to provide services, like banking and transport ticketing. Similar to smart cards, NFC-based transactions are susceptible to relay attacks. Distance bounding protocols have been proposed for smart cards to counter relay attacks. However, this may not be effective in the field of mobile transactions, due to their requirement of high time-delay sensitivity and specialised hardware. A number of proposals are being put forward that show that sensing the natural ambient environment is an effective anti-relay mechanism. Existing literature neither involves a threat actor in their analysis nor they are in compliance with EMVs transaction requirement of 500ms. In this paper, we look at the anti-relay mechanism from a different point of view. Instead of measuring the natural ambience, we generate and measure a unique artificial ambient environment (AAE) using peripherals of the devices involved in a transaction. To evaluate our proposal and its effectiveness, we selected infrared from the proposed set of off-the-shelf actuator/sensor pairs available on modern smartphones. We designed and deployed six distinct test-beds, each based on a unique method of relay attack, in order to evaluate the effectiveness of our proposal in the context of infrared. From our experimentations, we can empirically state that infrared showed high success rate in relay attack detection - higher than any existing work in academic literature.

Evaluation of Apple iDevice Sensors as a Potential Relay Attack Countermeasure for Apple Pay

Traditional countermeasures to relay attacks are difficult to implement on mobile devices due to hardware limitations. Establishing proximity of a payment device and terminal is the central notion of most relay attack countermeasures, and mobile devices offer new and exciting possibilities in this area of research. One such possibility is the use of on-board sensors to measure ambient data at both the payment device and terminal, with a comparison made to ascertain whether the device and terminal are in close proximity. This project focuses on the iPhone, specifically the iPhone 6S, and the potential use of its sensors to both establish proximity to a payment terminal and protect Apple Pay against relay attacks. The iPhone contains 12 sensors in total, but constraints introduced by payment schemes mean only 5 were deemed suitable to be used for this study. A series of mock transactions and relay attack attempts are enacted using an iOS application written specifically for this study. Sensor data is recorded, and then analysed to ascertain its accuracy and suitability for both proximity detection and relay attack countermeasures.

On the Effectiveness of Ambient Sensing for Detecting NFC Relay Attacks

Smartphones with Near-Field Communication (NFC) may emulate contactless smart cards, which has resulted in the deployment of various access control, transportation and payment services, such as Google Pay and Apple Pay. Like contactless cards, however, NFC-based smartphone transactions are susceptible to relay attacks, and ambient sensing has been suggested as a potential countermeasure. In this study, we empirically evaluate the suitability of ambient sensors as a proximity detection mechanism for smartphone-based transactions under EMV constraints. We underpin our study using sensing data collected from 17 sensors from an emulated relay attack test-bed to assess whether they can thwart such attacks effectively. Each sensor, where feasible, was used to record 350-400 legitimate and relay (illegitimate) contactless transactions at two different physical locations. Our analysis provides an empirical foundation upon which to determine the efficacy of ambient sensing for providing a strong anti-relay mechanism in security-sensitive applications. We demonstrate that no single, evaluated mobile ambient sensor is suitable for such critical applications under realistic deployment constraints.

Artificial Ambient Environments for Proximity Critical Applications

In the field of smartphones a number of proposals suggest that sensing the ambient environment can act as an effective anti-relay mechanism. However, existing literature is not compliant with industry standards (e.g. EMV and ITSO) that require transactions to complete within a certain time-frame (e.g. 500ms in the case of EMV contactless payments). In previous work the generation of an artificial ambient environment (AAE), and especially the use of infrared light as an AAE actuator was shown to have high success rate in relay attacks detection. In this paper we investigate the application of infrared as a relay attack detection technique in various scenarios, namely, contactless transactions (mobile payments, transportation ticketing, and physical access control), and continuous Two-Factor Authentication. Operating requirements and architectures are proposed for each scenario, while taking into account industry imposed performance requirements, where applicable. Protocols for integrating the solution into the aforementioned scenarios are being proposed, and formally verified. The impact on the performance is assessed through practical implementation. Proposed protocols are verified using Scyther, a formal mechanical verification tool. Finally, additional scenarios, in which this technique can be applied to prevent relay or other types of attacks, are discussed.

Extending EMV Tokenised Payments to Offline-Environments

Tokenisation has been adopted by the payment industry as a method to prevent Personal Account Number (PAN) compromise in EMV (Europay MasterCard Visa) transactions. The current architecture specified in EMV tokenisation requires online connectivity during transactions. However, it is not always possible to have online connectivity. We identify three main scenarios where fully offline transaction capability is considered to be beneficial for both merchants and consumers. Scenarios include making purchases in locations without online connectivity, when a reliable connection is not guaranteed, and when it is cheaper to carry out offline transactions due to higher communication/payment processing costs involved in online approvals. In this study, an offline contactless mobile payment protocol based on EMV tokenisation is proposed. The aim of the protocol is to address the challenge of providing secure offline transaction capability when there is no online connectivity on either the mobile or the terminal. The solution also provides end-to-end encryption to provide additional security for transaction data other than the token. The protocol is analysed against protocol objectives and we discuss how the protocol can be extended to prevent token relay attacks. The proposed solution is subjected to mechanical formal analysis using Scyther. Finally, we implement the protocol and obtain performance measurements.

Log Your Car: The Non-invasive Vehicle Forensics

Digital forensics is becoming an important feature for many embedded devices. In automotive systems, digital forensics involves multiple electronic control units (ECUs) used to support the connected and intelligent vehicles technology. Digital evidence from these ECUs can be used in forensics investigation and analysis. Such a mechanism can potentially facilitate crash investigation, insurance claims and crime investigation. Issues related to forensics include the authenticity, integrity and privacy of the data. In this paper, the security of the forensic process and data in automotive systems is analysed. We propose an efficient, secure, privacy-preserving and reliable mechanism to provide a forensics data collection and storage process. A diagnostic application for smart phones, DiaLOG, is incorporated in the proposed process that uses a secure protocol to communicate the collected forensic data to a secure cloud storage. The proposed protocol for communicating forensic data is implemented to measure performance results and formally analysed using Scyther and CasperFDR with no known attack found.

Reprint of “You can’t touch this: Consumer-centric android application repackaging detection”

Abstract Application repackaging is a widely used method for malware distribution, revenue stealing and piracy. Repackaged applications are modified versions of original applications, that can potentially target large audiences based on the original application’s popularity. In this paper, we propose an approach for detecting repackaged applications. Our approach takes advantage of the attacker’s reluctance to significantly alter the elements that characterise an application without notably impacting the application’s distribution. These elements include the application’s name and icon. The detection is initiated from the client side, prior to an application’s installation, making it application store agnostic. Our experimental results show that detection based on our algorithm is effective and efficient.

May the Force Be with You: Force-Based Relay Attack Detection

Relay attacks pose a significant threat against communicating devices that are required to operate within a short-distance from each other and a restricted time frame. In the field of smart cards, distance bounding protocols have been proposed as an effective countermeasure, whereas, in the field of smartphones, many proposals suggest the use of (natural) ambient sensing as an effective alternative. However, empirical evaluation of the proposals carried out in existing literature has reported negative results in using natural ambient sensing in distance- and time-restricted scenarios, like EMV contactless payments that require the proximity to be less than 3 cm and the transaction duration to be under 500 ms. In this paper, we propose a novel approach for Proximity and Relay Attack Detection (PRAD), using bidirectional sensing and comparing button presses and releases behaviour (duration of press and gap between presses and releases), performed by a genuine user during the transaction. We implemented a test-bed environment to collect training and analysis data from a set of users, for both the genuine and attacker-involved transactions. Analysis of the collection-data indicates a high effectiveness of the proposed solution, as it was successful in distinguishing between proximity and relay-attack transactions, using thresholds set after analysis of genuine training transaction data. Furthermore, perfect classification of genuine and relay-attack transactions was achieved by using well-known machine learning classifiers.