Dynamic Network Slicing Using Deep Reinforcement Learning
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | http://hdl.handle.net/10362/155750 |
Resumo: | Nowadays network slicing is one of the biggest drivers of new elements in the 5G network business. This is because this paradigm allows the creation of independent slices, with their virtually and logically separated radio, network and computational resources. Using network slicing, operators sell infrastructure resources of any kind to tenants, while tenants use these resources to sell services to their customers, the end users. In this context, a problem that is essential to solve is how to improve the operator’s profit, ensuring compliance with the requests’ SLAs and distributing network resources in order to increase its usage rate. This dissertation proposes to design two algorithms based on DRL for slice admission in the transport network, learning which request to accept and reject while guaranteeing the requirements of the tenants requests. The contributions of this study start with the formalization of the problem of slice admission, followed by its simulation and implementation of DRL agents using Containernet, the Ryu controller, OpenAI Gym and the PyTorch framework. The result is two DRL-based algorithms capable of achieving good performances in this simulated scenario. |
id |
RCAP_b4bac802ec20c926b2e5ec7912a91505 |
---|---|
oai_identifier_str |
oai:run.unl.pt:10362/155750 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Dynamic Network Slicing Using Deep Reinforcement LearningNetwork slicingDeep Reinforcement LearningAdmission ControlDomínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e InformáticaNowadays network slicing is one of the biggest drivers of new elements in the 5G network business. This is because this paradigm allows the creation of independent slices, with their virtually and logically separated radio, network and computational resources. Using network slicing, operators sell infrastructure resources of any kind to tenants, while tenants use these resources to sell services to their customers, the end users. In this context, a problem that is essential to solve is how to improve the operator’s profit, ensuring compliance with the requests’ SLAs and distributing network resources in order to increase its usage rate. This dissertation proposes to design two algorithms based on DRL for slice admission in the transport network, learning which request to accept and reject while guaranteeing the requirements of the tenants requests. The contributions of this study start with the formalization of the problem of slice admission, followed by its simulation and implementation of DRL agents using Containernet, the Ryu controller, OpenAI Gym and the PyTorch framework. The result is two DRL-based algorithms capable of achieving good performances in this simulated scenario.Atualmente o network slicing é um dos maiores potenciadores de novos elementos no negócio das redes 5G. Isto deve-se ao facto de este paradigma permitir a criação de slices independentes, com os seus recursos rádio, de rede e computacionais virtual e logicamente separados. Utilizando network slicing, as operadoras poderão vender recursos de infraestrutura de qualquer tipo a tenants. Os tenants utilizam estes recursos para vender serviços aos seus clientes, os utilizadores finais. Neste contexto, um problema que é fundamental resolver é o de como melhorar o lucro da operadora, garantindo o cumprimento dos SLAs dos pedidos e distribuindo os recursos da rede de forma a aumentar a sua utilização. Nesta dissertação propõe-se desenhar dois algoritmos baseados em DRL para a admissão de slices na rede de transporte, aprendendo que pedidos aceitar e rejeitar, procurando satisfazer sempre os requisitos dos pedidos dos tenants. Os contributos deste estudo passam pela formalização do problema da admissão de slices na rede, seguindo-se a sua simulação e implementação dos agentes utilizando conjuntamente o Containernet, o controlador Ryu, o OpenAI Gym e o framework PyTorch. O resultado são dois algoritmos baseados em DRL capazes de atingir boas performances neste cenário simulado.Amaral, PedroRUNOliveira, Pedro Miguel Silvério de2023-07-24T14:58:41Z2022-022022-02-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10362/155750enginfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2024-03-11T05:38:21Zoai:run.unl.pt:10362/155750Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-20T03:56:12.382256Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Dynamic Network Slicing Using Deep Reinforcement Learning |
title |
Dynamic Network Slicing Using Deep Reinforcement Learning |
spellingShingle |
Dynamic Network Slicing Using Deep Reinforcement Learning Oliveira, Pedro Miguel Silvério de Network slicing Deep Reinforcement Learning Admission Control Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática |
title_short |
Dynamic Network Slicing Using Deep Reinforcement Learning |
title_full |
Dynamic Network Slicing Using Deep Reinforcement Learning |
title_fullStr |
Dynamic Network Slicing Using Deep Reinforcement Learning |
title_full_unstemmed |
Dynamic Network Slicing Using Deep Reinforcement Learning |
title_sort |
Dynamic Network Slicing Using Deep Reinforcement Learning |
author |
Oliveira, Pedro Miguel Silvério de |
author_facet |
Oliveira, Pedro Miguel Silvério de |
author_role |
author |
dc.contributor.none.fl_str_mv |
Amaral, Pedro RUN |
dc.contributor.author.fl_str_mv |
Oliveira, Pedro Miguel Silvério de |
dc.subject.por.fl_str_mv |
Network slicing Deep Reinforcement Learning Admission Control Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática |
topic |
Network slicing Deep Reinforcement Learning Admission Control Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática |
description |
Nowadays network slicing is one of the biggest drivers of new elements in the 5G network business. This is because this paradigm allows the creation of independent slices, with their virtually and logically separated radio, network and computational resources. Using network slicing, operators sell infrastructure resources of any kind to tenants, while tenants use these resources to sell services to their customers, the end users. In this context, a problem that is essential to solve is how to improve the operator’s profit, ensuring compliance with the requests’ SLAs and distributing network resources in order to increase its usage rate. This dissertation proposes to design two algorithms based on DRL for slice admission in the transport network, learning which request to accept and reject while guaranteeing the requirements of the tenants requests. The contributions of this study start with the formalization of the problem of slice admission, followed by its simulation and implementation of DRL agents using Containernet, the Ryu controller, OpenAI Gym and the PyTorch framework. The result is two DRL-based algorithms capable of achieving good performances in this simulated scenario. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-02 2022-02-01T00:00:00Z 2023-07-24T14:58:41Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10362/155750 |
url |
http://hdl.handle.net/10362/155750 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799138148022747136 |