.png){kind=icon}
.png){kind=icon}
Return
WiLabel: Behavior-Based Room Type Automatic Annotation for Indoor Floorplan
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2019.2922842.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The growing indoor location based services (LBS) applications enhance the requirement of room type annotation. Existing room type annotations are either depending on additional sensors or prone to privacy disclosure. We proposed a method called WiLabel-based on channel state information (CSI) alone. By analyzing the CSI fluctuation, we adopt the percentage of nonzero elements (PEM) algorithm to classify indoor scene and design a zero prior knowledge behavior recognition method to achieve behavior perception in the fewer-person scene, then, design a behavior-based decision tree classifier to determine the room type. The evaluation results from 84 rooms of the college building and mall show that the WiLabel can achieve an average accuracy of 90.5% superior to others.
Keywords:
Channel state information
indoor floorplan
room type annotation
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.