Institutional Repository of Vilnius University
Metaeuristinių algoritmų taikymas energiją taupančiam realaus laiko užduočių planavimui daiktų interneto sistemose /
Abstract
dc:descriptionThis paper examines energy saving in real-time periodic task scheduling of IoT systems. In the paper, it was studied whether task offset optimization can be implemented for such a purpose in a way that would save energy and guarantee meeting task deadlines. In this work, a periodic task model, an energy model and a simulation environment were implemented. It implemented the classical rate-monotonic scheduler and earliest deadline first algorithms, an energy saving procrastination heuristic, as well as three metaheuristic algorithms - genetic algorithm, simulated annealing, particle swarm optimization. Metaheuristic algorithms were used as a way to get better optimized offset vector values. Experiments tested the effects of different tasksets, processor utilization levels, energy profiles and metaheuristic algorithm parameter configurations. The results show, that metaheuristic algorithms can improve the results in asynchronic or offset free systems. However, in the case of synchronic systems, metaheuristic algorithms only managed to reach, and rarely slightly surpass the results of the classical algorithms. Procrastination heuristic substantially beat every other method discussed in this paper. The results show, that optimizing just the offsets is not enough to clearly surpass the energy saving heuristic. Still, this optimization may be useful in systems that already have offsets and it is required to minimize the impact they have on energy usage.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Bacius, Domantas,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD308335981&prefLang=en_US