Modelagem matemática do atraso de entrega de mensagens em redes oportunistas com taxas de encontro heterogêneas

• Ano de defesa: 2018
• Autor(a) principal: Dias, Gabriela Moutinho de Souza
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Rio de Janeiro; Brasil; Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia; Programa de Pós-Graduação em Engenharia de Sistemas e Computação; UFRJ
• Palavras-chave em Português: Modelagem matemática; Redes Oportunistas; Multi-copy; Heterogeneidade; CNPQ::ENGENHARIAS
• Link de acesso: http://hdl.handle.net/11422/13150

Presents the development of two mathematical models to capture the expected end-to-end message delivery delay in opportunistic networks, for singlecopy and multi-copy forwarding process, when the source node can generate multiple copies of the message. The research area of opportunistic networks is becoming stronger over the years. However, despite of the diversity of protocols and solutions proposed for dealing with the challenges of this environment, there is an open issue in the literature that is the lack of general mathematical models. This gap is the major motivation for the present work. The mathematical modeling proposed in this document focus on the dynamics of encounters among nodes, taking into consideration the heterogeneity in their mobility, which assumes that the pairwise encounter rates are different. The first model was inspired by a single-copy model of the literature and expanded to the multi-copy case. The second model, which considers both single and multi-copy cases, follows a different approach and was developed to eliminate some approximations used in the first model. Both models were validated by comparing analytical and simulation results. Three simulation tools were used: a proper simulator specially developed for this work and the network simulators ns-3 and The ONE. The obtained results show the high precision of the estimates, for both synthetic and real life traces.