SUMMARY of the Trust4All project

 Trustworthiness in embedded software

The aim of the project:
The Trust4All project aims at the development of a middleware software architecture specifically targeted at embedded systems that require a defined level of trust, due to the nature of the services they provide. The growing importance of software in embedded systems, its economic value and the fact that individuals and societies depend more and more on the correct functioning of these embedded systems are the major reasons behind the initiative. From a technology point of view the manageability of the embedded software in systems is ever growing due to the change from closed, not connected static systems to open, inter-connected and dynamic systems. This offers all kind of opportunities for the usability of these systems; on the other hand it also gives rise to all kind of threats w.r.t. the trustworthiness of the embedded software behaviour. As such the project will focus on the trustworthiness related aspects of the middleware software architecture in application domains that require a certain level of dependability as the are home medicare, home security, domotica, and on-the-move applications.

The need for dependability:

When analysing the future needs of the envisaged application domains from a user point of view it's clear that there is an increased expectation on the expected level of dependability these system can provide. These dependencies will require defined levels (ranging from high to medium) of expected trustworthiness of the system, namely: · Security during critical actions to guarantee the expected behaviour, · Reliability during its lifetime to secure the expected usefulness, · Robustness during operation to guarantee the expected functionality.

The goals of Trust4All: 

The project intends to develop a trust model driven by the requirements of the targeted application domains. The model will allow the precise specification of the nature and level of the expected level of dependability and should be translatable into requirements and constrains on the embedded software composition. A methodology to validate if a certain (dynamic changeable) composition of embedded software satisfies the requirements needs to be developed to allow for the deployment of the trust model. The software architecture will be open with published interfaces and methods supporting emerging ecosystems for third-party middleware in the embedded software domain. Especially for this purpose the project intends to submit its results to standardisation bodies such as the ISO MPEG activity on middleware.

Trust4All challenges:

The main challenge of the project is the definition of a trust model that can be applied in the anticipated application domains and, if possible, can be applied in related domains as well. This requires the ability do deal with a wide variety of requirements at different levels of severity. This implies that (application domain specific) metrics need to be defined that requires an extensive study of the application domain. Care should be taken that the approach used for the metrics, as well as for the trust model, need to be such that the application domain agnostic nature of the approaches is secured to allow for the deployment in different applications as well. Secondly there is the challenge to validate the level of dependability of a composed system, with respect to a desired state, in a dynamically changing setting. Also for this challenge the application domain agnostic nature of the approach needs to be safeguarded.

How to tackle the challenge:

The project will start with an analysis related to the trust requirements in the envisaged application domains with the intention do define the trust model. This model will be validated with experts form these application domains. The next step is the development of an architecture and associated life cycle model with embedded validation mechanisms supporting checking and / or enforcement of a required trust level at defined intervals or events. All these developments will be promoted to international standardisation of guarantee wide applicability and availability.

Trust4All expected result(s):

The major result of the project is the trust model with an architecture that provides the ability to validate, at any stage in time, the level of trust of a composed system against the model. As such the most important deliverables of the project are a trust model with associated metrics and a methodology to (dynamically) validate the status of a system with respect to this model after (requested or un-requested) updates, upgrades or extensions. On the basis of the outcome of this validation process a system is allowed or not to execute a certain application or service, requiring a defined level of trust.

The Consortium:
  1. Philips Research Laboratories, established in The Netherlands
  2. Philips Applied Technologies, established in The Netherlands
  3. Technische Universiteit Eindhoven, established in The Netherlands
  4. LIACS – University of Leiden, established in The Netherlands
  5. Océ Technologies B.V., established in The Netherlands
  6. Stichting Telematica Instituut, established in The Netherlands
  7. Centrum voor Wiskunde en Informatica, established in The Netherlands
  8. VTT Technical Research Centre of Finland, established in Finland
  9. Nokia Corporation, established in Finland
  10. Solid Information Technology, established in Finland
  11. Fundacion Robotiker, established in Spain
  12. Fagor Electrodomesticos, S. Coop., established in Spain
  13. Ikerlan, established in Spain
  14. Visual Tools, established in Spain
  15. Fundacion European Software Institute, established in Spain
Project start July 1, 2005 

Project end September 30, 2007 

"The work of Spanish partners has been partially funded by the National Programme on Information Technologies of the Ministry of Industry, Trade and Tourism under grant FIT-340000-2006-62".

You need more information ?

In that case please contact  the Trust4All
for editorial issues please contact

[Return to homepage]
Last updated on 13 Aug 2008   (c) Trust4All project