부산대학교 도서관

Large networks of fixed or mobile sensors based on Internet of Things technologies generate a large amount of data, that requires specific methods for preprocessing, averaging, interpolation and graphic representation. In this, bidimensional scalar data are taken into account; we consider the issues of window averaging, interpolation, and convex hull definition. The considered scenario is a fully automated, near real-time acquisition and visualisation system. The aim of this work is to compare the quality and the computational cost of various interpolation methods, while highlighting the possible risks behind the need to introduce experts' choices in some of them. In this, some algorithmic optimisations and comparisons are presented, based on both simulated and real data. We present some software solutions for the problems of window averaging and buffering of the convex hull and some tests on interpolation methods like Inverse Distance Weighting and Natural Neighbours to determine which is the most appropriate for our real-time measurements. [ABSTRACT FROM AUTHOR]