Spatial data analysis for intelligent buildings: Awareness of context and data uncertainty

Abstract

Intelligent buildings are among the most active Internet-of-Things (IoT) verticals, encompassing various IoT-enabled devices and sensing technologies for digital transformation. Analysis of spatial data, a very common type of data collected in intelligent buildings, offers a lot of insights for many purposes such as facilitating space management and enhancing the utilization efficiency of buildings. In this paper, we recognize two major challenges in spatial data analysis for intelligent buildings (SDAIB): (1) the complicated analytical contexts that are related to the building space and internal entities and (2) the uncertainty of spatial data due to the limitations of positioning and other sensing technologies. To address these challenges, we identify and categorize different kinds of analytical contexts and spatial data uncertainties in SDAIB, and propose a unified modeling framework for handling them. Furthermore, we showcase how the proposed framework and the associated modeling techniques are used to enable context-aware and uncertainty-aware SDAIB, in the tasks of hotspot discovery, path planning, semantic trajectory generation, and distance monitoring. Finally, we offer several research directions of SDAIB, in line with the emerging trends of the IoT.