Towards a Security Domain Model for Embedded Systems

Abstract

The need for efficient processes for implementing security properties in systems that are made of networked embedded devices motivates a fresh look at how security is "added" to embedded components at late development stages. In this short paper we propose a security domain model intended for considering security aspects at early embedded systems design stages. It consists of two components: (1) An interface to the system engineering models in which aspects that are relevant to security are extracted, and (2) Elements that are specific to known security solutions and realisable in collections or libraries. The paper shows that what this security domain model has in common with other software security models is the need for representation of security properties, and what it has in common with other embedded systems models is the need for representation of resources. The proposed model is described and illustrated by application to a mesh communication for crisis management scenario.