FAST HEURISTIC SCHEDULING BASED ON NEURAL NETWORKS FOR REAL-TIME SYSTEMS

As most of the real-time scheduling problems are known as hard problem s, approximate or heuristic scheduling approaches are extremely requir ed for solving these problems. This paper presents a new heuristic sch eduling approach based on a modified Hopfield-Tank neural network to s chedule tasks with deadlines and resource requirements in a multiproce ssor system. In this approach, fast heuristic scheduling is achieved b y performing a heuristic scheduling policy in conjunction with backtra cking on the neural network. The results from our previous work, using the same neural network architecture without backtracking, are includ ed here as a case with zero backtracking. Extensive simulation, which includes comparison with the conventional heuristic approach, is used to validate the effectiveness of our approach.