Uma abordagem cross-layer para redes corporais sem fio considerando dispositivos com taxas de dados variáveis

• Ano de defesa: 2019
• Autor(a) principal: Rogério Santos Pozza
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ENG - DEPARTAMENTO DE ENGENHARIA ELÉTRICA; Programa de Pós-Graduação em Engenharia Elétrica; UFMG
• Palavras-chave em Português: Redes corporais sem fio; Cross-layer; Monitoramento corporal; WBAN; Engenharia elétrica; Postura humana
• Link de acesso: http://hdl.handle.net/1843/34855

Research on wireless body area networks (WBAN) has gained interest and has become an emerging theme of health services due to its potential use in the continuous monitoring of human well-being. The natural movements of the human body and environmental interference impose difficulties on the performance of a WBAN, such as a lifetime and reliability of data delivery. Due to the complexity of the WBANS characteristics, it is a challenge to maintain the quality of the links among the network sensor nodes while extending the battery life. Different experimental studies in the literature show that the use of fixed and high power transmission may not be adequate, although it allows a more reliable communication, it can lead to unnecessary energy consumption. On the other hand, low power transmission offers more energy savings but may reduce the reliability of packet delivery. Research on data transmission in WBANs is essential as an energy-saving strategy, as power control techniques allow sufficient power levels to be selected to achieve excellent performance and further reduce interference problems. This work proposes a cross-layer approach, called CL, based on the IEEE\-802.15.4 standard to ensure the quality of service (QoS) in terms of packet delivery and network lifetime. The CL approach proposes to share parameter values between protocol stack layers to improve efficiency in sink packet delivery and power consumption. The results from this work were obtained through the implementation of WBANs with the Castalia simulator. It was used three scenarios, with different data rates and two types of communication channels (with and without interference). The results comparison of the scenarios concerned packet delivery, power consumption, and communication media adaptability. Three cross-layer approaches proved this proposition: CL-2, CL-4, and CL-Bin. CL-2 uses two levels of Tx power, CL-4 four levels, where the choice is by the number of packets received by the sink. CL-Bin is a predictive approach using a probabilistic model to determine the power of Tx. The results of these approaches, compared to a conventional layered WBAN, show that the CL approach is more efficient when considering the proposed scenarios.