Modelagem e simulação do deslocamento de pessoas para estimativa de formação de grupos

• Ano de defesa: 2013
• Autor(a) principal: Véras, Frank César Lopes
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Informática; Programa de Pós-Graduação em Informática; UFPB
• Palavras-chave em Português: Rede de Sensores sem Fio; Wireless sensors network; ZigBee; Simulação; Ptolemy II; Radio Frequency Identification - RFID; Formação de grupos; Simulation; CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/tede/7823

The use of Wireless Sensor Networks (WSN) has been widespread in many areas of research and application. The purpose of this work is a study on how to use sensors to monitor people on the move, having a WSN as a way in which the search will occur, but in order to predict the formation of groups in certain regions. For this work the network was designed and tested in the simulator Ptolemy II, using the ZigBee Communication protocol, where the sensors were positioned according to a Cartesian coordinate system. The WSN will detect people and identify common patterns of movement, such as speed, direction and type of movement, using parameters set in the simulator. People involved in the groups will be identified by RFID (Radio Frequency Identification) attached to his body. The movement of the crowd had its mathematical formalization based on parameters such as position of the group, number of people per group and duration of movement that define the characteristics necessary to simulate this scenario. From the formalization of the movement of the crowd, many data are collected at predetermined time intervals and interpreted by an algorithm, through the exchange of messages between sensors, estimates the crowd forming in the region defined as the target. In this work, were inserted charts and graphs that reflect the actual number of people moving towards the real target. These data are generated from the intense exchange of messages between sensors, obeying some parameters that favor established and the algorithm that estimates the crowd at the target formation at any given time. The accuracy of the prediction was measured by the amount of alarms issued that estimate and the formation of agglomerations of people in a given region. Thus, the identification of individuals by sensors is interpreted according to the possibility of formation of groups and their values disseminated by the network. The proposal is that this action will facilitate the process of decision making and thus help to characterize the formation of crowds.