Related Experiment Video
Updated: Feb 12, 2026

Building Double-decker Traps for Early Detection of Emerald Ash Borer
Published on: October 4, 2017
Double-Windows-Based Motion Recognition in Multi-Floor Buildings Assisted by a Built-In Barometer
Maolin Liu1, Huaiyu Li2,3, Yuan Wang4
1Institute of Remote Sensing and GIS, Peking University, No. 5 Yiheyuan Road, Haidian District, Beijing 100871, China. maolin@pku.edu.cn.
Smartphone barometers can now accurately recognize vertical motion, like using stairs or elevators, achieving 95.05% accuracy. This new method improves recognition by analyzing pressure signals and handling imbalanced data effectively.
Area of Science:
- Sensor fusion and signal processing for human motion recognition.
- Mobile sensing and ubiquitous computing applications.
- Machine learning for pattern recognition in time-series data.
Background:
- Traditional smartphone sensors (accelerometers, gyroscopes, magnetometers) have limited accuracy for vertical motion detection.
- Barometers offer improved vertical motion sensing but lack quantitative signal analysis and modeling.
- Imbalanced data presents a challenge for accurate motion recognition algorithms.
Purpose of the Study:
- To develop and validate a method for accurate vertical motion recognition using smartphone barometers.
- To address the limitations of existing sensors and data imbalance issues in motion recognition.
- To provide a robust strategy for enhancing vertical motion detection in multi-floor environments.
Main Methods:
- Modeling and feature extraction of barometer pressure signals using a novel double-windows approach.
- Implementation of a random forest classifier with a correlation rule to mitigate imbalanced data effects.
- Quantitative analysis of pressure signals for distinguishing vertical movements like stair climbing and elevator use.
Main Results:
- Achieved a recognition accuracy of 95.05% for vertical motion recognition using barometer data and an improved random forest classifier.
- Significantly enhanced accuracy and response time for recognizing stair and elevator movements.
- Demonstrated the effectiveness of the proposed double-windows feature extraction and classifier in improving vertical motion detection.
Conclusions:
- Smartphone barometer-based vertical motion recognition is highly accurate and effective, particularly for multi-floor navigation.
- The proposed methods successfully address challenges of sensor limitations and imbalanced datasets.
- This work provides a robust and accurate strategy for improving vertical motion recognition, enhancing mobile sensing capabilities.
Related Concept Videos
Definition and Measurement of Pressure: Atmospheric Pressure, Barometer, and Manometer
Muscles of the Pelvic Floor and Perineum
Perineal Layer
The perineum is a diamond-shaped area below the pelvic diaphragm, divided into an anterior urogenital triangle that contains the external genitals and a posterior anal triangle housing the anus. The urogenital...
Movement Joints in Buildings
The simplest type of movement joints, working joints, are...
Types of Building Stone
Igneous rocks are formed from the solidification of magma or lava. An example is granite, known for its durability and resistance to weathering, making it ideal for parts of...
Equation of Motion: General Plane motion
Moreover, the body's center of mass experiences a rotational effect as a result of these couple moments. This rotation can be articulated as the...
Fixing Double-strand Breaks

