Information Technology Reference
In-Depth Information
the project. Therefore SQUARE does not rely on a pre-defined terminology that
we could use.
3.3 State of the Art of Security-Oriented Modelling Languages
Many security modelling languages, or most often security extensions to exist-
ing languages, were developed. Existing approaches based on UML have been
enriched with security modelling capabilities. In Misuse Cases [ 51] and Abuse
Cases [ 42] , which are extensions of “Use Case” diagrams, the focus is on elici-
tation of new threats and vulnerabilities that could be exploited by malicious actors.
SecureUML [ 35] extends several UML diagrams. The approach focuses on authori-
sation constraints and its goal is to automatically generate complete access control
infrastructures. UMLsec [ 28] is a UML profile that allows adding security-related
information to UML diagrams. Both SecureUML and UMLsec address security at
the design level. They, thus, do not focus on business assets and high-level security
requirements.
The KAOS goal-oriented framework addresses security concerns by treating
attacks as anti-goals [ 29] . Anti-goals are the attacker's goals and generate obstacles
to security goals. Extensions of the i goal-oriented framework [ 57] also address
security problems. For instance, Liu et al. [ 34] represent attacks as tasks with neg-
ative contributions to security softgoals. A formalisation of i to deal with security
issues is proposed in Secure-Tropos [ 17, 47] . It suggests, first, to extend the concepts
and the processes of i /Tropos and, then, to integrate techniques such as security
reference diagrams and security attack scenarios. Recently, additional work [ 12]
has been done on representing the notion of vulnerability in i . Asnar et al. intro-
duced the Tropos Goal-Risk Framework [ 2] that addresses RM at three different
levels, combining together asset, risk, and risk treatment views. However, the Tropos
Goal-Risk framework does not focus on IS security, but supports the concept of
risk in general, including project management risk and financial risk, for instance.
Finally, Problem Frames extensions were also proposed to handle security issues.
Anti-requirements were introduced by Abuse Frames [33] . Abuse Frames are used
to delimit the scope of a security problem and thereby are meant to facilitate the
analysis of threats and vulnerabilities as well as the elicitation of security require-
ments. In future work, we plan to confront the concepts of these languages with the
concepts of the ISSRM domain.
4 ISSRM Concept Alignment
4.1 Concepts to Consider
The first task of the concept alignment phase is to define the range of concepts
to study. In [14] , a comparison between the concepts used in various security RE
 
 
Search WWH ::




Custom Search