A special issue of the Journal of Computer Security (IOS Press, ISSN 0926-227X) brings together the research results of six ongoing FP6-IST projects. The editors of the journal invited those projects that received outstanding reviews in the ICT Trust and Security EU-calls to submit their papers.
Security and trust are core research issues for the further development of the information society and for ten years have played, and continue to play, an integral part in the European Union's Framework Programmes (FPs) for R&D. EU-supported collaborative research projects in trust and security bring together multi-partner stakeholders from industry (technology and service providers, system integrators and end-users), academic and research laboratories working in several interdisciplinary research fields. Sometimes, projects include actors from the legal, social and economic sectors. Together their joint efforts permit a better understanding of the conflicts and synergies between security, privacy and free market economics, as well as of the psychology and sociology of trust and security when building and deploying new technologies. The long term goal for funding this research effort is to convert the know-how of the EU in security, privacy and trust into economic advantages.
In the period 1998-2002, under FP5, original and ground-breaking scientific & technological (S&T) work took place in ICT Trust and Security. Key S&T developments achieved at that period included significant advances in cryptology, smart cards and biometrics. EU-supported research also permitted the identification of new concepts in fields of work such as privacy, dependability and risk analysis. In the period 2002-2006, under FP6, research efforts in ICT security and trust have been further intensified. As part of the FP6-IST Programme, 37 R&D projects have been funded, with a total value of €239 million (€140 million funding from the European Commission).
About the papers
"New filtering approaches for phishing email", by the IST project ANTIPHISH (www.antiphishresearch.org). Today's state-of-the-art anti-phishing approaches are not broadly implemented. They lack completion or require a high administrative overhead. In this paper, ANTIPHISH Partners present some new and effective solutions addressing e-mail phishing. They concentrate on countermeasures based on the contents of phishing e-mails and describe a number of novel features that are particularly well-suited to identify phishing e-mails. The range of proposed features include: statistical models for the descriptions of e-mail topics, sequential analysis of e-mail texts and external links contained in an e-mail, detection of embedded logos, and indicators for intentional addition or distortion of content not perceivable by the reader. The new solutions have been tested using a large real-life corpus of e-mails pre-labelled as spam, phishing, and legitimate. The results for the classifications of phishing versus legitimate and phishing versus legitimate and spam indicate that the authors achieve very low error rates and outperform other schemes previously proposed for classifying phishing e-mails.
"Provably correct inline monitoring for multithreaded java-like programs", by the IST project S3MS (www.s3ms.org/index.jsp). Today's mobile devices have the computing power to offer users many more applications than currently available. S3MS developed the 'security-by-contract' paradigm, which essentially consists of the use of policies, monitoring, and monitor inlining to secure third-party mistrusted applications running on mobile devices. Inlining is a program rewriting technique to ensure that a program complies with a given security policy. The paper presents the design and implementation of inlined reference monitors. More precisely, a security state is embedded into the client program and code rewriting is used in order to ensure that such a state is queried and updated appropriately. This solution presents some very original features: firstly, it allows for automatic enforcement of a previously defined policy; and secondly, since the policy is formulated in a 'contract' (i.e., a piece of code) that accompanies the program, the policy is easily available for consultation to interested parties.
"Security of trusted repeater quantum key distribution networks", by the IST project SECOQC (www.secoqc.net). Modern cryptography relies on the use of digital 'keys' to encrypt data before sending it over a network, and to decrypt it at the other end. The receiver must have a version of the key code used by the sender so as to be able to decrypt and access the data. Quantum key distribution (QKD) offers a theoretically uncrackable code, one that is easily distributed and works in a transparent manner. Even better, the nature of quantum mechanics means that if any eavesdropper tries to snoop on a message, the sender and receiver will both know. The SECOQC project has investigated the feasibility of an open QKD infrastructure, together with the enabling technology. The aim was to enhance the existing landscape of security technologies with novel and reliably secure ways of performing long-range and high-rate distribution of secret keys. The paper proposes a provably secure method using the QKD protocols in a trusted key repeater network. The deployment of such a network is feasible with current technology (ordinary computers and optical networks) and will most likely be the way quantum key-distribution will be employed in the real world.
"Towards automated security policy enforcement in multi-tenant virtual data centers", by the IST project OPENTC (www.opentc.net). Securing our computing systems and making them trustworthy is one of themost challenging problems in computer science for the years to come. The OPENTC project aimed to improve the security of computing infrastructures by combining virtualization and trusted computing using open-source software. Virtual systems allow for fine-grained isolation of confine attacks as well as errors. Trusted computing enables stakeholders to validate the configuration of given systems. The use of open source components allows users to examine, validate, and improve upon the functionality of the trusted computing base. Exemplifying that at a specific scenario, the paper presents a security architecture for virtual data centres based on Trusted Computing technologies. The architecture allows for the automatic deployment of the security mechanisms required by the policy that was specified for the particular service. The authors have shown the viability of their approach by a reference implementation.
"Biometric template protection in multimodal authentication systems based on error correcting codes", by the IST project HUMABIO (www.humabio-eu.org). HUMABIO was concerned with human monitoring and authentication using biodynamic indicators and behavioural analysis. Biometric authentication, particularly at airports, is one of the most widespread and visible security mechanisms. The paper demonstrates an advanced stride forward in establishing and improving such authentication mechanisms. It proposes a framework for authentication combining data from multiple biometric sensors, e.g., facial and gait properties. The proposed solution not only improves the identification process, but also protects the privacy of users by hiding their original templates using cryptography. The underlying idea is to consider the biometric recognition as a distributed source-coding problem to which well-known error correcting codes can be applied. The authors have shown the viability of their system with tests at an airport.
About Journal of Computer Security
The Journal of Computer Security presents research and development results of lasting significance in the theory, design, implementation, analysis, and application of secure computer systems. It also provides a forum for ideas about the meaning and implications of security and privacy, particularly those with important consequences for the technical community.
The special issue on EU-Funded ICT Research on Trust and Security is available as of now (Journal of Computer Security 18:1, 2010) and is guest-edited by Jan Camenisch, Javier Lopez, Fabio Massacci, Massimo Ciscato and Thomas Skordas.