Security

Our work in this area is concerned with the development and precise definition of modelling languages, methodologies and ontologies to support elicitation, modelling and analysis of security, trust, and privacy requirements. We have developed the Secure Tropos methodology, one of the first methodologies in the literature that implements the idea of security and privacy by design and integrates security, privacy and engineering techniques under one methodological approach.

Our work in that area focuses on the development and analysis of methods, processes, and architectures for secure and privacy-aware systems. At the requirements level, we focus on the development of processes that enable the elicitation and modelling of security and privacy requirements and analyse them in terms of security and privacy properties, relevant threats and vulnerabilities. At the architectural level, we focus on developing software architecture techniques to ensure that software systems satisfy security, trust and privacy requirements and that developed architectures reduce potential risks. Such techniques include model-based methodologies that enable support at different levels, from design –through the development of architectural models, to run time – through the execution and management of such models. Our work also includes traceability support between software architecture and other artefacts of the software lifecycle such as code and requirements.

Recent advances in information and communication technologies such as Internet of Things, 5G and Cloud Computing, provide significant benefits but at the same time raise important issues with regards to security and privacy. We are developing novel models, methodologies and analysis techniques that guarantee the highest possible levels of protection within IoT, 5G and Cloud computing environments, in the presence of different security and privacy threats.

One of the single largest concerns facing organisations today is how to protect themselves from cyber-attacks whose prominence impose the need to prioritise their cyber security concerns with respect to their perceived threats. We are investigating novel decision-making methodologies and models that offer the highest possible levels of protection within different domains (e.g. IoT, Cloud) with regards to different security and privacy threats and a set of evolving factors such as security requirements, financial cost, indirect costs (e.g. people’s productivity), intangible and tangible assets. We also develop underlying formalisms, utilising logics and graph transformations, to enable precise specifications and automated reasoning, within the context of security and dependability, taking into account organisational policies and resource allocation.

We develop novel reasoning techniques and algorithms that assist the discovery of potential cyber-attack paths in supply-chain and critical infrastructures, taking into account information from the Common Weakness Enumeration (CWE) and from the Common Vulnerabilities and Exposures (CVE). Our work can be applied within a dynamic risk management system to detect the vulnerabilities of the IT infrastructure and to deliver attack paths that satisfy certain criteria.

We develop platforms and solutions that facilitate visual analysis of privacy requirements and needs and assist the creation, monitoring and enforcement of Privacy Level Agreements. Moreover, we focus our work in the analysis and development of innovative data privacy governance platforms, which facilitate scoping and processing of data and data breach management and support organisations towards GDPR compliance.

Security and privacy patterns capture the experiences of experts, allowing novices to rely on expert knowledge and solve security and privacy problems in a more systematic and structured way. The literature provides catalogues of security and privacy patterns. However, representing and selecting security and privacy patterns remains largely an empirical task. Our work contributes towards the solution of this problem in two ways. On one hand, we investigate the development of security and privacy pattern languages that enable the representation of patterns and guide developers through the process of designing a system to ensure security and privacy. A major novelty of our work in that area is that the solution to the pattern is represented using concepts from the requirements stage, which enable a developer to directly apply the patterns of the language to the security and privacy requirements analysis. We also pioneer work on security and privacy patterns selection, which allows us to understand in depth the trade-offs involved in patterns and the implications of a pattern to various security and privacy requirements. The process then supports the search for a combination of security and privacy patterns that will meet given security and privacy requirements.

To support software development process activities, we are developing computer-aided software engineering (CASE) tools. Our tools support security and trust analysis of the system under development at different levels. At the higher level, they are graphical editors where security and trust models can be drawn and the grammatical correctness of the models is automatically checked. On the lower level, they enable analysis of security and trust properties and security threats.