ENF-S: An Evolutionary-Neuro-Fuzzy Multi-Objective Task Scheduler for Heterogeneous Multi-core Processors

In this paper, an online task scheduling and mapping method based on a fuzzy neural network (FNN) learned by evolutionary multi-objective algorithm (NSGA-II) to jointly optimize the main design challenges of heterogeneous MPSoCs is proposed. approach, first, FNN parameters are trained using NSGA-II-based optimization engine considering including temperature, power consumption, failure rate, execution time training dataset consisting different application graphs various sizes. Next, employed as scheduler in MPSoCs. Due uncertainty sensor measurements difference between computational models reality, applying advantageous procedures. The performance compared with some previous heuristic, meta-heuristic, rule-based approaches several experiments. Based these experiments our proposed outperforms related studies optimizing all criteria. Its improvement over heuristic meta-heuristic estimated 10.58% 9.22% 39.14% 12.06% time, averagely. Moreover, interpretable nature FNN, frequently fired extracted rules approach demonstrated.