Modeling of Legacy Communication in Component-Based Distributed Embedded Systems Saad Mubeen * , Jukka M¨ aki-Turja *† and Mikael Sj¨ odin * * M¨ alardalen Real-Time Research Centre, M¨ alardalen University, V¨ aster˚ as, Sweden † Arcticus Systems, J¨ arf¨ alla, Sweden {saad.mubeen, jukka.maki-turja, mikael.sjodin}@mdh.se Abstract We propose the addition of special purpose component types to a commercially existing component model, the Rubus Component Model (RCM). The purpose of the new component types is to encapsulate and abstract the communications protocol and configuration in a component based and model based software engineer- ing setting. With the addition of these new component types, RCM will be able to support state-of-the-practice development processes of distributed embedded systems where communication rules are defined early in the development process. We also show how an end-to- end timing model can be extracted from a distributed embedded system, modeled with RCM, to perform end- to-end timing analysis. Keywords Model-Based Software Engineering; Component-Based Software Engineering; Distributed embedded systems. I. Introduction With the recent advancement in technology, embedded systems have become more and more complex. In order to deal with this complexity, lower development cost, reduce time-to-market and time-to-test, allow reusability and support modeling at higher level of abstraction, etc., the research community proposed the employment of Model-Based Engineering (MBE) and Component-Based Software Engineering (CBSE) for the development of embedded systems [1] [2]. Software development of distributed embedded sys- tems is more complex as compared to single processor embedded systems. When MBE and CBD are used for the development of resource constrained and hard real- time distributed embedded systems, modeling of commu- nication infrastructure arises as another challenge. The component model for the development of such systems should not only be resource efficient but it should also abstract the application software from the communication infrastructure. Moreover, it should also be able to model the legacy (previously developed) communications and legacy systems. In this paper we propose the extension of a commer- cially existing component model, the Rubus Component Model (RCM) [3], by adding special purpose component types to it. RCM is a component model used for the development of resource constraint real-time embedded systems. It supports glue-code generation, end-to-end delay analysis, and resource requirements estimations. The purpose of the new component types, introduced in RCM, is to encapsulate and abstract the communications protocol and configuration in a component based and model based software engineering setting. Our main goals in introducing these components are: 1) Allow model-based and component-based develop- ment of new nodes that are deployed in legacy systems that use predefined communications rules. 2) Support state-of-the-practice development pro- cesses where communications rules are defined early in the development process. 3) Enable adaptation of a node when communications rule change (e.g. due to re-deployment in a new system or due to upgrades in the communication system) without affecting the internal component design. These goals are to be realized in RCM. The scope of this paper is PSMs (Platform Specific Models) for distributed embedded systems. With PSM we mean that the software component has been allocated to nodes and any adaptation to specific node characteristics (e.g. device drivers and memory layouts) has been added to the model. Using our new components, nodes can be developed without explicit knowledge about the communication configuration. Paper Layout The rest of the paper is organized as follows. Section II presents the Rubus concept, the component model and its development environment. In section III, we present the related research and compare different modeling