Towards Automatic Spatial Verification of Sensor Placement in Buildings

Abstract

Most large, commercial buildings contain thousands of sensors that are manually deployed and managed. These sensors are used by software and firmware processes to analyze and control building operations. Many such processes rely on sensor placement information in order to perform correctly. However, as buildings evolve and building subsystems grow and change, managing placement information becomes burdensome and error-prone. An automatic verification process is needed. We investigate empirical methods to automate spatial verification. We find that a spatial clustering algorithm is able to classify relative sensor locations -for 15 sensors, spread across five rooms in a buildingwith 93.3% accuracy, 13% better than a k-means clusteringbased baseline method. Analysis on the raw time series data has a classification accuracy of only 53%. By decomposing the signal into intrinsic modes and performing correlation analysis, an observable, statistical boundary emerges that corresponds to a physical one. These results may suggest that automatic verification of placement information is possible.

Automated Metadata Construction to Support Portable Building Applications

“…There have been various data-driven efforts to capture the contextual and semantic relationships between sensors in order to build applications, such as type classification of sensors [19] and finding spatial relationships between sensors ( [11,15,17]). However, these techniques either classify sensors into broad categories (type classification), and do not capture semantic (or functional) relations between sensors (e.g which air temperature sensor is related to which setpoint sensor), and hence are not useful in writing applications which depend on the semantic relationships between sensors.…”

“…Request permissions from Permissions@acm.org. BuildSys' 15 or SCADA (Supervisory Control and Data Acquisition) systems that host higher level control, retain historical data, and provide visualization. Many of these systems provide some kind of programmatic interface to the sensors, actuators, and historical data under their management [2,7,24].…”

Comparison of linear correlation and a statistical dependency measure for inferring spatial relation of temperature sensors in buildings

“…The main idea behind inferring room connectivity is based on the similarity of motion and/or light sensor firings. There has also been some effort put into classifying relative sensor locations by using the data collected from those sensors, which is also the main objective of this study [2]. The main goal is to find a statistical boundary between sensor data correlations that can be taken as an indicator of spatial relation.…”

“…For example, the importance of location metadata for each sensor in the building has been previously discussed in the literature [1]. On the other hand, keeping metadata updated can be a formidable effort, since it is likely that sensors are replaced over time and/or the layout of the building changes during the building's life [2]. For instance, layout renovations are frequent and occur in 30% of all spaces annually due to maintenance purposes or changes in space requirements [3].…”

“…Inferring relative sensor locations in a building based on the linear correlation coefficients between datasets collected from these sensors has also been studied [2]. However, none of these studies employed a non-linear correlation measure to accomplish sensor localization.…”