Simulating a multi-vehicle traffic sensing system based on radio tomographic imaging

• Ano de defesa: 2016
• Autor(a) principal: Wilkens, Jarmo Theodore
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Engenharia Elétrica; Centro Tecnológico; UFES; Programa de Pós-Graduação em Engenharia Elétrica
• Palavras-chave em Português: Imagens tomográficas de rádio (RTI: Radio Tomographic Imaging); Serviços baseados em localização; Redes de sensores sem fio; Pessoas - Localização; Sistemas inteligentes de veículos rodoviários; Engenharia Elétrica; 621.3
• Link de acesso: http://repositorio.ufes.br/handle/10/9672

With the increasing amount of vehicles on the road, traffic jams pose a growing global problem. Traffic surveillance is a crucial task into improving traffic flow. The core of this dissertation proposes a design and methodology for the real-time estimation of occupancy and velocity of one or two vehicles on a single lane road segment, traveling at speeds from 10 km/h up to 50 km/h, using a Radio Tomographic Imaging (RTI) network. RTI is an emerging technology and consists of a wireless sensor network that produces images of the change in the electromagnetic field of a monitored area using Received Signal Strength (RSS) measurements. This allows the tracking of device-free objects such as humans and cars. The proposal is the first to present an RTI network that is split into multiple sub-networks to achieve a network scan frequency fast enough to prevent blurring in the captured images of vehicles traveling at velocities higher than 50 km/h. This proposal suggests three newly implemented techniques, namely a new weight matrix calculation method, a new node spacing setup and a new vehicle detection method. The main contribution of this dissertation is to stimulate research of the possibility of using RTI networks as being part of an Intelligent Transport System (ITS) in a smart city environment. This dissertation also presents a new, relatively complete simulation model that simulates the proposed RTI system. It generates RSS measurements based on the IEEE 802.15.4 protocol for ZigBee communication, which are processed using RTI techniques commonly used in literature. The simulation model allows the specification of an RTI network from the ground up, including node positions, network size and geometry and RSS measurement processing using any available or newly developed techniques. The simulation results, obtained after implementing the proposed RTI system to the validated simulation model, indicate that it is possible to detect both one or two family sized cars simultaneously. Using techniques that reduce RSS variance due to multipath effects and the newly proposed methods, simulated vehicle detection performance is demonstrated to be between 95% and 100%.