Platform-Specific Code Generation from Platform-Independent Timed Models

Abstract

Many safety-critical real-time embedded systems need to meet stringent timing constraints such as preserving delay bounds between input and output events. In model-based development, a system is often implemented by using a code generator to automatically generate source code from system models, and integrating the generated source code with a platform. It is challenging to guarantee that the implemented systems preserve required timing constraints, because the timed behavior of the source code and the platform is closely intertwined. In this paper, we address this challenge by proposing a model transformation approach for the code generation. Our approach compensates the platform-processing delays by adjusting the timing parameters in system models, based on an Integer Linear Programming problem formulation. We demonstrate the usefulness of our approach via a case study of infusion pump systems. Experimental results show that the code generated using our approach can better preserve the timing constraints. Disciplines Computer Engineering | Computer Sciences Comments

Determining Timing Parameters for the Code Generation from Platform-Independent Timed Models

“…This framework checks if the original parameters of the platform-independent models can be used to generate the code, but does not change the parameters in case the platform-specific model does not verify the timing requirements. The second approach was to systematically adjust such timing parameters of the platform-independent model so that the generated code can be integrated with the platform in a way that meets the timing requirements (Kim et al 2015b). In comparison to those prior works, the main contribution of this article is to propose a unified framework that allows one to apply those model transformations under different conditions to determine the timing parameters for the code generation purposes.…”

“…In general, the computation of the two functions is difficult due to the complex dependency between the clock variables and I/O transition, and the computation algorithm may not terminate in the presence of the cycles (Kang and Lee 1996). However, if we restrict the timed automata to have the clock reset on every transition, f M can be computed by applying the shortest and longest path algorithm in the graph theory (e.g., Floyd-Warshall algorithm) and detailed in our prior work (Kim et al 2015b). All parameters to compute f M are known a priori such as the automata structure (e.g., acyclic or cyclic automata) and the timing parameters.…”

“…Under a certain restricted timed automata structure, the delay bounds for all pairs of transitions can be formulated by considering only simple paths. More details on the extension can be found in our prior work (Kim et al 2015b).…”

“…We repeated this measurement 20 times from which the min/max platform delays were obtained. The details of the model and the measured platform delays can be found in Kim et al (2015b). Given the two inputs (i.e., the platform-independent model and the platform delays), we formulated the linear constraints according to the method presented in Section 5 to find the new timing parameters for the code generation.…”

“…Informally, this constraint states the min/max timing bound of input/output that a system will produce since an input (output) has been received (produced) from (to) the environment. We extend our two prior works (Kim et al 2015a(Kim et al , 2015b to propose a unified code generation framework in order for the implemented system to meet the delay-bound inclusion constraint. Given a platform chosen to execute the code, this extension enables one to incorporate the platform delays to the platform-independent model through two step transformations.…”