Platform-Specific Timing Verification Framework in Model-Based Implementation

Abstract

In the model-based implementation methodology, the timed behavior of the software is typically modeled independently of the platform-specific timing semantics such as the delay due to scheduling or I/O handling. Although this approach helps to reduce the complexity of the model, it leads to timing gaps between the model and its implementation. This paper proposes a platform-specific timing verification framework that can be used to formally verify the timed behavior of an implementation that has been developed from a platformindependent model. We first describe a way to categorize the interactions among the software, a platform, and the environment in the form of implementation schemes. We then present an algorithm that systematically transforms a platform-independent model into a platform-specific model under a given implementation scheme. This transformation algorithm ensures that the timed behavior of the platform-specific model is close to that of the corresponding implementation. Our case study of an infusion pump system shows that the measured timing delay of the system is bounded by the formally verified bound of its platform-specific model. Keywordsformal specification, program verification, scheduling, software metrics, timing, I/O handling, formal verification, infusion pump system, model complexity, model-based implementation methodology, platformindependent model, platform-specific model, platform-specific timing semantics, platform-specific timing verification framework, scheduling, timing delay, Automata, Computational modeling, Delays, Semantics, Software, Synchronization Abstract-In the model-based implementation methodology, the timed behavior of the software is typically modeled independently of the platform-specific timing semantics such as the delay due to scheduling or I/O handling. Although this approach helps to reduce the complexity of the model, it leads to timing gaps between the model and its implementation. This paper proposes a platform-specific timing verification framework that can be used to formally verify the timed behavior of an implementation that has been developed from a platform-independent model. We first describe a way to categorize the interactions among the software, a platform, and the environment in the form of implementation schemes. We then present an algorithm that systematically transforms a platform-independent model into a platform-specific model under a given implementation scheme. This transformation algorithm ensures that the timed behavior of the platform-specific model is close to that of the corresponding implementation. Our case study of an infusion pump system shows that the measured timing delay of the system is bounded by the formally verified bound of its platform-specific model.

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

“…We previously proposed two independent approaches. The first approach was to systematically extend the platform-independent model into the platform-specific model that is to be verified against the timing requirements (Kim et al 2015a). 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.…”

“…After completing those steps, the code becomes idle waiting for another invocation. algorithms to perform such syntactic construction are explained in our prior work (Kim et al 2015a). Figure 11 illustrates the timing semantics of the platform-specific model in comparison to the platform-independent model.…”

“…Proof. Refer to our prior work (Kim et al 2015a). If the implemented system is built according to the io-boundary interaction type, the actual delay of the implemented system is bounded by the calculated mc-delay from Lemma 4.3.…”

“…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.…”

Platform-Specific Code Generation from Platform-Independent Timed Models

“…In our previous work, we proposed to overcome this obstacle by modeling architectural aspects of platforms [12] [10] and generating test cases [11]. However, one limitation is that, the code generated from models that incorporate platformprocessing delays, when being implemented on a specific platform, may yield a system that does not respect timing constraints verified at the modeling level.…”