The latest forecasts on the increase in connected devices (IoT paradigm) and the traffic exchanged among them, stress the need for the efficient management of the resources available in wireless networks. The location of mobile devices is an essential information to achieve that goal. While most of the devices are able to obtain their position, however such a process tends to kill the device in terms of power requirements and network traffic. This can be even worse when considering that all the connected devices need to constantly update such information. Moreover, the current trend of diversifying and densifying the wireless infrastructure in order to absorb the expected growth in traffic, boosts the possibility of geo-positioning solutions: improving accuracy, latency, robustness of the system, etc. However, this network redundancy generates two clear challenges: 1) the management of an overwhelming amount of data in devices and network equipment, and 2) the management of the network infrastructure in an efficient and flexible way, in order to minimize the impact on energy waste and CO2 emissions. The current project aims at addressing some of these open issues in an area that has not yet been explored, thus constituting an important challenge and, at the same time, having enormous potential in the field of research. This project proposes, in the first place, the design of collaborative positioning algorithms for mobile devices, suitable for their use in areas where GPS cannot be used (e.g. indoors, urban canyons, devices powered by batteries, etc.). These algorithms have the triple objective of 1) maximizing the quality of service, 2) minimizing the energy consumption and 3) maximizing the scalability of the location system. The coordinated optimization in this scenario is important, since the three objectives cited tend to be detrimental to each other: maximizing efficiency, for example, means increasing device consumption and, often, limiting the scalability of the system. Secondly, the concept of location middleware is used to deal with the massive and constant positioning of IoT devices, the management of data and its accessibility from the Internet (Internet of Data). Machine learning algorithms will be applied to estimate the present and future positions of the network nodes, as well as the data mining algorithms to generate social and/or economic knowledge, which can be turned into new services or applied to the optimization of the network itself. Third, the location of IoT nodes can be used to study a geocasting solution that combines minimum latency in the delivery of data, robustness to changes in the network topology, low error rate and minimum power consumption. In addition, the data generated after applying the data mining processes will be used to identify the individual and group mobility patterns of the network devices to optimize the provision of services. Lastly, this project proposes the creation of a positioning platform, in which the solutions presented in the project can be implemented and evaluated.
Plan Estatal de Investigación Científica y Técnica y de Innovación 2017-2020
Programa Estatal de Generación de Conocimiento y Fortalecimiento Científico y Tecnológico del Sistema de I+D+i
Subprograma Estatal de Generación de Conocimiento
Proyectos de I+D de generación de conocimiento (antigues EXC)