A SIMULATION SCHEME FOR ESTIMATING DEADLINE OF REAL-TIME TASK MODELEDIN TIMED PETRI-NET

Estimating the deadline of a real-lime task is a necessary prerequisit e to the applications that have strict timing constraints, such as rea l-time systems design. This paper shows how Monte-Carlo simulation can be used as a space-efficient way of analyzing Timed Petri nets to pre dict whether the system specified tan satisfy its real-time deadlines. For the purpose, Extended Timed Petri Net (XTPN), an extension of con ventional Timed Petri net. and its execution rule, using Monte-Carlo t echnique, are newly defined. A simple simulation scheme with less memo ry space is presented as a way of estimating the deadline of a real-ti me task modeled in XTPN. And the comparison between the analytical and simulation results is given, The problem addressed here is to find th e probabilities of meeting given deadlines.