Thematic Group 3: Security, Dependability and Trust

by Michel Riguidel

ICT security balances freedom and the desire to protect tangible and intangible assets, ensures the immunity of applications and system resilience, and instils and preserves confidence in digital, critical infrastructures. At the smallest level, nanotechnology, quantum communication and cryptography offer new opportunities to tackle ICT security.

Embedded sensors and devices can form ad-hoc networks requiring new mechanisms for establishing trust when sharing information or resources. New paradigms come to the foreground, such as service architectures that compose services from lower level modules, peer-to- peer systems characterized by their remarkable robustness and resilience against attack, and biological defence mechanisms which may inspire new breakthrough technologies. At a larger scale, the completion of the Galileo satellite navigation system around 2009 will create ever more sophisticated possibilities for positioning with implications for both security and privacy.

Proposed Research Themes
In view of the above, the following emerging research themes were identified by experts participating in this Thematic Group:

1. Ambient Trustworthiness
The mass diffusion of digital systems must be enabled with built-in mechanisms for enhancing trust and confidence on their usage. Common security mechanisms mainly based on boundaries and firewall protection mechanisms do not scale with respect to new complex systems. We should imagine different mechanisms such as the ones inspired by the living world: immune and self-healing systems. We should consider autonomic, evolvable and adaptive security mechanisms, which will require new semantic models managing the complexity of ambient intelligence environments where humans and devices may jointly function and interact. Security systems and cryptographic mechanisms must be scaled down for inclusion in small devices (even at nano-scale) with specific requirements for energy consumption and computation power.

2. Dynamicity of Trust
Lack of trust either on the cyber-infrastructure (due to frequent attacks) or the difficulties to model trust relationshipsamong different entities (human and digital ones) is one of the main barriers for the establishment of a true Information Society. In future ICT systems with billions of interconnected devices, the capability of managing and negotiating trust relationships that foster cooperation is crucial. The understanding of how trust emerges and evolves, as well as of the related notions of reputation formation, monitoring and evolution are mandatory. Security-based trust and trust-based security are two emerging areas of interest. A deeper understanding of trust needs the involvement of research expertise from several fields such as economy and sociology.

3. Quantum Technology and Cryptography
Nature can provide us with resources to secure our information and communication systems. The possibility provided by Quantum technology to offer secret bits of information among authenticated distant partners, as well as truly random values, could serve as building blocks of many protection mechanisms. Quantum technology and quantum computing might also represent a major threat for current cryptographic algorithms and mechanisms. We should also study the assumptions under which Quantum Computers (QC) may act and their consequences on current and future cryptographic methods, as well as the development of new QC resistant cryptographic techniques.

4. Assessability and Verifiability
Assessing and proving the trustworthiness of a complex system is a major issue. During the last years many techniques have been developed, especially in the dependability community. Yet, the scale of new ICT systems and the kind of threats and assumptions about their operational environment pose new challenges. Different metrics, modelling tools and observation mechanisms are needed. The capability of measuring the tolerance to attacks is crucial in new systems that due to their logical and physical diffusion are susceptible to 'attack'. There is a need to develop a discipline of software and system security based on high-level verifiably secure programming. This calls for an approach based on efficiently verifiable mathematical proofs showing compliance to policies (expressing safety, security, or functionality constraints).

Participate in the online consultation of this report from 1 February to 31 March 2006 at

TG3 Coordinator:
Michel Riguidel, ENST-Paris, France