| Περίληψη: | This diploma thesis describes the design and implementation of a Relative
Indoor Localization Algorithm. A lot known indoor localization systems require
prior knowledge of the environment in form of statistic data, multiple sensors
(odometry on moving targets) or directional antennas to work reliably with high
accuracy. The innovation of this thesis comes from the diversification provided by
the Relative part of the localization system, as opposed to the absolute
localization algorithms available. A great number of different situations does not
require pinpoint accuracy in locating the objects that need to be localized. Instead
an indication on the position of the object, in relation to other known locations is
closer to the human’s localization approach, e.g. “somewhere in the living room”.
This kind of model provides all the information needed for a human to interpret
the data and extract a result even if it lacks accuracy. This is an ideal candidate for
e-health monitoring due to its non-invasive nature and the fact that there is
minimum setup to be able to extract results. Throughout this thesis there will be a
thorough analysis of the algorithm and the architecture developed for this
purpose, as well as a presentation of the hardware and software used for the
implementation. Ending this thesis, will follow a presentation the results and a
complete analysis of the known issues as well as some discussion containing
potential use cases.
|
|---|
