Efficient end-to-end latency distribution analysis for probabilistic time-triggered systems

Medical image processing systems are typically implemented with a number of independent subsystems that have time-triggered interfaces. A critical design parameter for these systems is the latency between the instant that an image is captured and the instant that the enhanced image is displayed to the physician. Computation of the end-to-end latency distribution with existing techniques is often impractical due to the extremely large number of states that need to be considered. In this paper we introduce the probabilistic time-triggered system (PTTS) model. With this model the end-to-end latency distribution can be computed efficiently. Furthermore, we present a bound on the time-complexity of our analysis algorithm and a technique that reduces the state-space at the cost of accuracy. We demonstrate the applicability of the presented analysis technique by showing that several system configurations of an X-Ray application can be quickly explored. This exploration reveals the parameters which have a significant effect on the end-to-end latency distribution.