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

Baekgyu Kim, Lu Feng, Oleg Sokolsky, Insup Lee

Research output: Contribution to journalArticlepeer-review

Abstract

Safety-critical embedded systems often need to meet dependability requirements such as strict input/output timing constraints. To meet the timing requirements, the code generation (e.g., C code) from timed models needs to determine the timing parameters that indicatewhen the code has to perform I/Owith its platform.We propose a novel framework to determine such timing parameters from platform-independent timed models. Our framework involves two transformations. The first transformation systematically extends the platformindependent model by explicitly modeling input/output processing (e.g., sampling or interrupt-based) and the code invocation (e.g., periodic or aperiodic) mechanisms. Then, we verify if the resulting platform-specific model meets the timing requirements. In the case that the resulting model does not satisfy the timing requirements, we apply the second transformation to compensate the platform delay via adjusting the timing parameters at the code level. We formulate the adjustment mechanism using integer linear programming. If such an adjustment is feasible, generating the code with the new timing parameters guarantees the implemented system to meet the timing requirements. We validate our framework with case studies running on Patient-Controlled Analgesia (PCA) infusion pump platforms.

Original languageEnglish
Article number3230711
JournalACM Transactions on Cyber-Physical Systems
Volume3
Issue number3
DOIs
StatePublished - Aug 2019

Bibliographical note

Publisher Copyright:
© 2019 Association for Computing Machinery.

Keywords

  • Code generation
  • Medical device
  • Platform-independent models
  • Platform-specific models
  • Real-Time embedded systems
  • Timed automata

Fingerprint

Dive into the research topics of 'Determining Timing Parameters for the Code Generation from Platform-Independent Timed Models'. Together they form a unique fingerprint.

Cite this