Modern vehicles such as cars, ships, and planes are increasingly managed using Electronic Control Units (ECUs) that communicate over a Controller Area Network (CAN) bus. While this approach offers enhanced functionality, efficiency, and robustness, it may also be used for unforeseen or malicious purposes ranging from aftermarket modifications to full-fledged attacks threatening passengers’ safety. The ability to conduct in-depth tests is thus vital to protect against these issues. However, much of the functionality of ECUs is proprietary or undocumented. To alleviate this obstacle, this work presents a reverse engineering approach using high-coverage test sets produced using Combinatorial Testing (CT) methods. Our results indicate that this technique is promising for exciting unknown functionality, although challenges regarding the presence of hidden state and high-accuracy oracles are yet to be overcome.

Summary This work presents an extended and enhanced gray‐box combinatorial security testing methodology for SQL injection vulnerabilities in web applications. We propose multiple new attack grammars modelling SQLi attacks against MySQL‐compatible databases, each one targeting a different injection context. Additionally, these grammars are also dynamically refined at the beginning of each attack against an endpoint of a web application, as a further optimization of the used attack model by taking into account the specifics of the generated query of that endpoint. Our goal is to enhance existing combinatorial approaches for detecting SQL injection vulnerabilities. The newly developed methodology is implemented in a prototype security testing tool called SQLInjector+, which is an extension of an earlier prototype developed by us in prior work. This improved tool can attack (i.e. test) any web application that uses a MySQL‐compatible database management system. We evaluate our revised approach and improved prototype tool in a case study comprising of different kinds of web applications to which SQLi is a potential security threat. The case study contains the well‐known verification framework WAVSEP among other five real‐world web applications and one web application firewall. Our generated attack vectors, constructed via combinatorial methods applied to our improved and dynamically optimized attack grammars, are capable of injecting every known vulnerable endpoint in WAVSEP and also of finding new vulnerable parameters in some of the real‐world applications investigated in this paper. Our approach performs equally well or better when compared with existing state‐of‐art of SQL injection security testing tools (sqlmap, w3af, wapiti and fuzzdb) across all tested web applications in the case study.

Current web crawlers, document databases and change monitoring systems for web sites are commonly limited to static content and analysis of code as retrieved from the server, an approach that is not suitable for modern dynamic web applications. The canonical representation of the contents of a single web page at any given time is an instance of the Document Object Model (DOM), a tree structure that forms the basis for rendering and processing of the page within the browser and is updated when content is modified. This work presents DOMdiff, an algorithm to identify changes between two different DOM instances, as well as a method to classify these changes in terms of a ranking that represents the distance between the two trees. We compare a manually derived classifier with the results of PRank, a ranked version of the Perceptron algorithm, a simple machine learning approach that generates a multiclass classifier based on formulae in a constrained predicate logic, and the established statistical classifier C5.0. Our results indicate that DOMdiff is suitable to large-scale change identification and that entropy-based statistical classifiers are more accurate than our simple predicate-based classifier for the problem at hand, but require a larger decision tree. We additionally identify a shortcoming of PRank when handling features with low information gain/high entropy.

As malicious software gets increasingly sophisticated and resilient to detection, new concepts for the identification of malicious behavior are developed by academia and industry alike. While today's malware detectors primarily focus on syntactical analysis (i.e., signatures of malware samples), the concept of semantic-aware malware detection has recently been proposed. Here, the classification is based on models that represent the underlying machine and map the effects of instructions on the hardware. In this paper, we demonstrate the incompleteness of these models and highlight the threat of malware, which exploits the gap between model and machine to stay undetectable. To this end, we introduce a novel concept we call covert computation, which implements functionality in side effects of microprocessors. For instance, the flags register can be used to calculate basic arithmetical and logical operations. Our paper shows how this technique could be used by malware authors to hide malicious code in a harmless-looking program. Furthermore, we demonstrate the resilience of covert computation against semantic-aware malware scanners.

Recently, academia and law enforcement alike have shown a strong demand for data that is collected from online social networks. In this work, we present a novel method for harvesting such data from social networking websites. Our approach uses a hybrid system that is based on a custom add-on for social networks in combination with a web crawling component. The datasets that our tool collects contain profile information (user data, private messages, photos, etc.) and associated meta-data (internal timestamps and unique identifiers). These social snapshots are significant for security research and in the field of digital forensics. We implemented a prototype for Facebook and evaluated our system on a number of human volunteers. We show the feasibility and efficiency of our approach and its advantages in contrast to traditional techniques that rely on application-specific web crawling and parsing. Furthermore, we investigate different use-cases of our tool that include consensual application and the use of sniffed authentication cookies. Finally, we contribute to the research community by publishing our implementation as an open-source project.

This paper examines QR Codes and how they can be used to attack both human interaction and automated systems. As the encoded information is intended to be machine readable only, a human cannot distinguish between a valid and a maliciously manipulated QR code. While humans might fall for phishing attacks, automated readers are most likely vulnerable to SQL injections and command injections. Our contribution consists of an analysis of the QR Code as an attack vector, showing different attack strategies from the attackers point of view and exploring their possible consequences.