Architecture for multi-technology real-time location systems

• Autores: Javier Rodas González
• Directores de la Tesis: Carlos J. Escudero (dir. tes.)
• Lectura: En la Universidade da Coruña ( España ) en 2015
• Idioma: inglés
• Número de páginas: 183
• Tribunal Calificador de la Tesis: Mónica Fernández Bugallo (presid.) , Paula-María Castro-Castro (secret.) , Joaquín Miguez Arenas (voc.)
• Enlaces
  • Tesis en acceso abierto en: RUC
• Resumen

  • Indoor localization is a problem that has generated much interest in recent years. Proximity marketing, eHealth, smart-parking and smart-cities, security and emergency units, logistics management, or industrial control systems are some pf the sectors that have demanded new Location Based Services (LBSs). These services are usually implemented using Wireless Sensor Networks (WSNs), capable of transmitting and receiving Radio Frequency (RF) signals in order to locate mobile devices attached to vehicles, people, or animals. While systems based on satellite systems such as GPS work correctly in outdoor scenarios, indoor localization is still a challenging field of study. On one hand, signal propagation problems are common, not only due to reflections and scattering due to the building structures, but also because of signal attenuation and fading caused mainly by people in movement. To overcome these issues, most of the approaches use several WSNs with a combination of multiple wireless technologies, such asWiFi, ZigBee or Bluetooth, some of them also available in mobile devices such as smartphones and tablets. On the other hand, data received from multiple devices must be filtered and combined by means of location algorithms and techniques in order to obtain precise and robust Real-Time Location Systems (RTLSs). Therefore, it is common to implement hybrid location systems with support for several technologies at the same time. Nevertheless, the development of such systems entails a huge complexity. Thus, one of most widely accepted alternatives is the implementation of software architectures for localization, which provide several benefits. First, accessing to different kinds of hardware devices entails fewer platform and technology restrictions. Second, some common tasks are easier to perform, such as sensor data gathering and storage. Finally, architectures provide utilities for adding and retrieving localization data, user management, or the possibility of using several mapping and coordinate systems. In this work, we present several solutions for implementing software architectures for localization. First, we propose a mono-technology architecture using only Received Signal Strength (RSS) signal levels for ranging, which evolves into a much more complete multitechnology architecture in a second stage. The proposed approaches implement several functionalities that resolve most of the hybrid RTLS system requirements, such as: - Multi-technology. - Support for several coordinate systems and mapping applications. - Data fusion. - Protection and security for both data and user access. - Standardized API for remote access. - Support for off-line data queries, not only on-line data and in real-time. - Depending on different user roles, it eases their tasks at different access levels: registration of WSNs, building blueprints, anchor and mobile node networks registration, generic sensor support, addition and retrieval of measurements and raw sensor data, multiple query support for filtered position estimations, etc. Moreover, we also contributed with different WSN physical layer implementations and experiments. And, due to collaborations with other research groups at different universities we have contributed with a customized hardware and software solution for localization based on RFID technology, as well as with the design of new antenna models based on linear-arrays of Electromagnetically Coupled Patchs (ECPs), valid for improving the WSN communication performance.