A hybrid framework of worst-case execution time analysis for real-time embedded system software

Timing analysis^1,2 is an essential process for development of real-time embedded system and knowledge about the worst-case execution time (WCET) of real-time programs is critical to validation of temporal correctness of implemented system. Recently, automated static timing analysis methods are introduced to facilitate timing analysis process for real-time software, and to provide safe and tight WCET. But static WCET analysis methods have drawback as they do not provide accurate WCET for hardware-dependent software or application software where input data rate from external environment needs to be considered. Also, the WCET obtained from static WCET analysis needs to be verified at target system before system deployment. In this paper, we propose a framework of WCET analysis for real-time embedded software which complements static WCET approach and provides tight and safe WCET by combining static timing analysis approach with dynamic measurement. The application of proposed framework to the WCET analysis of command processing and data acquisition part of KOMPSAT-2 satellite flight software is presented to show effectiveness of the proposed approach.