Publication
Particionamento hardware/software: proposta de solução com redes neurais artificiais
| Summary: | The hardware/software partitioning problem is one of the most important step on project development when it has a hardware/software co-design. Partitioning problems are include in the set of NP-Complete problems and when they have a optimizing feature associated the are moved to the set of NP-Hard problems. This fact justifies the usage of heuristics on solving this problem. The choosen heuristics are genetic algorithms for database generation and artificial neural networks for problem solving. The result of this work is a neural network that can solve the problem for an especific instance with good results in a short comptational execution time. |
|---|---|
| Main Authors: | Dias, Maurício Acconcia |
| Subject: | Hardware/software co-design Particionamento hardware/software Redes neurais artificiais Hardware/software co-design Hardware/software partitioning Artificial neural networks |
| Year: | 2015 |
| Country: | Brazil |
| Document type: | bachelor thesis |
| Access type: | open access |
| Associated institution: | IBICT |
| Language: | Portuguese |
| Origin: | Oasisbr |
| Summary: | The hardware/software partitioning problem is one of the most important step on project development when it has a hardware/software co-design. Partitioning problems are include in the set of NP-Complete problems and when they have a optimizing feature associated the are moved to the set of NP-Hard problems. This fact justifies the usage of heuristics on solving this problem. The choosen heuristics are genetic algorithms for database generation and artificial neural networks for problem solving. The result of this work is a neural network that can solve the problem for an especific instance with good results in a short comptational execution time. |
|---|
Funded activities
No funded projects found.