Scheduling algorithms for meet row constraint

The problem of guaranteeing "meet row n in m" weakly hard real-time constraint is addressed in this paper. The paper first analyzes characters of μ-pattern and gives out the definition of "unsatisfied μ-pattern". Then, it enumerates and proves several properties of unsatisfied μ-pattern. According to the properties, length of unsatisfied μ-pattern can indicate whether the μ-pattern satisfies the meet row constraint, therefore, a scheduling algorithm named meet row algorithm is presented based on the properties. Meet row algorithm tries to find out the probabilities of tasks breaking constraints and increase task\´s priority in advance. Simulation results show that meet row algorithm is better than other scheduling algorithms for meet row constraints and can largely decrease the worst-case computation time of real-time tasks.