Catálogo de publicaciones - libros

Engineers are faced nowadays with the challenge of designing strongly distributed control systems, with complex interactions. There is little theory and tool support to address this recent challenge. Control engineering and telecom engineering have dedicated but unrelated techniques, each for their specific domain. In this paper, we explore an approach where we combine two complementary formal methods, with good tool support and industrial acceptance:SCADE/Lustre from the Control Engineering domain, and SDL, from the Telecom domain.

In this paper we describe our experiences in building tools for accelerating maintenance of existing large telecommunications software. We discuss how various maintenance activities can be accelerated by providing developers with the knowledge of the core scenarios of the system, which approximate the intended use cases. We present a static approach to extracting scenarios as source trajectories, byna vigating through the source code and capturing the source statements as events. We describe our tool for static capturing of scenarios. The possibilityof static capturing of the core scenarios and their representation as MSCs have benefits in retaining expertise about existing software, in training new personnel, in focusing understanding of legacysoft ware, performing architecture reviews, and in architecture analysis of existing systems. We believe that this approach can contribute to closing the gap between tool support for forward engineering in the so-called “green-field” projects, and maintenance of existing software.

This paper discusses how MSC-2000 could influence the Sequence Diagrams within UML 2.0, and whythe UML 1.x semantics is partly inadequate for what is needed in the area of sequence charts. Extracts of a possible UML meta-model is shown and this can be understood as a conceptual model for MSC-2000 as well and an indication of an approach to the future MSC-2000 semantics. UML Collaboration Diagrams have no direct counterpart in SDL/MSC and the difference between sequence diagrams and collaboration diagrams is analyzed.

Designing complex systems as stacks of collaborating layers is a common practice in various domains, from operating systems to user interfaces. It proves to be particularly fruitful in the domain of telecom systems and, more generally, in distributed systems. After showing why, today, SDL is not well suited for the design of layered systems, we explore usual techniques available in programming languages. Then, we attempt giving SDL the power of expression necessary to view and manipulate signals at different abstraction levels while preserving the language spirit, staying at design level.

The concept of capturing dynamic aspects of a distributed system across agent boundaries is elaborated in the context of SDL-2000. Several ways of composing collaborations are introduced, with collaborations being implicitly represented as SDL fragments. A new language for their formal description, called CoSDL (Collaborations in systems) is then introduced and illustrated.

The purpose of Implementation Design is to bridge the gap between an abstract system, e.g. in SDL, and its implementation in hardware and software. Expressing the Implementation Design decisions is a general challenge in systems engineering. Notations have been defined to describe abstract models and implementations, but little work has been done to define a notation for Implementation Design. In this paper, we discuss UMLsolutions and extend them. Our work is intended as a contribution to a common SDLapproac h, and an inspiration to ITU-T Study Group 10 question 11 on deployment and configuration language DCL. We also present how the ProgGen tool was extended in order to generate code controlled by the Implementation Design model.

The increasing complexity of SDL software (for example, as required for distributed system architectures and advanced operating system integrations) has generated a need for a powerful notation fordeplo yment. The notation should be used formo deling the run-time configuration of SDL applications and the communication between these. This paper formulates requirements on a deployment notation to be used with SDL. Using the mapping between SDL and UML in Z.109, it is investigated how the UML implementation diagrams can be used for deployment of SDL systems, and how well the diagrams conform to the requirements. It is found that the UML deployment diagram can be used forsho wing deployment of SDL agent instance sets and static instances. Agent instances dynamically created and destroyed require certain mapping rules in order to be modeled. The interface concept in UML is found to be adequate formo deling communication. Using the UML extension construct “tagged value”, information at arbitrary detail levels can be shown. This information can be used by a variety of targeting tools such as code generators.

This paper provides an introductory overview of ETSI testing activities related to interoperability and conformance testing. It explains why testing is important and why languages such as TTCN-3 are considered to be key components in the development of ETSI testing specifications.This paper highlights the relationship and harmonization between TTCN-3, MSC, SDL and ASN.1.

Experiences with the use of the MSC language for complex system specifications have shown that certain extensions are necessary in order to arrive at sufficiently transparent and manageable descriptions. Extended HMSCs, where MSC reference symbols may either be presented by hypertext-like descriptions or, in an expanded form, as detailed MSCs, appear to be especially suitable for a compact and transparent MSC representation. For an effective usage of such advanced MSC constructs, a corresponding tool support seems to be mandatory where interactively the event structures of special paths can explicitly be expanded while others remain hidden as MSC references that contain solely textual descriptions. The name is proposed for such extended HMSCs. Beyond that, the communication between MSC references, operator expressions or HMSCs demands a generalisation of the gate concept. For that purpose, the introduction of MSC connectors denoting logical connections is suggested. MSC may be expanded similar to MSC references. HyperMSCs enhanced by MSC connectors also provide a means for a selected visualisation of large MSCs in an interactive manner where, depending on the current selection, some parts are exhibited in full detail whereas other parts are presented in an abbreviated form. The same concepts may be applied for system modelling based on stepwise refinement starting with HyperMSCs, decomposed instances and MSC connector communication and for system testing.

Recently, the European Telecommunications Standards Institute (ETSI) approved the third edition of the Tree and Tabular Combined Notation (TTCN-3) as a requirement to modernise and widen its application beyond pure OSI conformance testing. As part of this evolution, TTCN is embracing Message Sequence Charts (MSCs) as a natural notation for specifying and visualising test suites. This paper defines the role of MSCs during test development, and more specifically introduces an MSC profile called the Graphical Format for TTCN (GFT) that facilitates the effective specification of TTCN-3 test suites.