Abstract
In recent year, as a major national strategic emerging industry, location based services (LBS) have been widely used and been affecting people's daily lives. With the speeding up development of urbanization and further improvement of people's living standard, the demand of three-dimensional location-based service grows rapidly when dealing with indoor navigation, asset management, security monitoring, security dispatching, emergency rescue, and many other aspects. However, at present the mainstream navigation and positioning system only provides users with two-dimensional position information that is accurate longitude and latitude, but not vertical positioning information because of the poor precision. This paper focuses on researching the differential pressure altimetry by comparing pressure altimetry. The differential pressure altimetry can provide accurate vertical height information for indoor and outdoor navigation and positioning system and will provide important theoretical basis and technical support for the evolution of traditional two-dimensional LBS for the high precision of three-dimensional indoor and outdoor seamless LBS.
Introduction
While modern navigation systems excel at providing two-dimensional coordinates, vertical accuracy remains a significant challenge. GPS/GNSS signals are often attenuated or unavailable in indoor environments, and standard barometric altimeters, though cost-effective, suffer from inherent instability. Atmospheric pressure is not solely a function of altitude; it fluctuates due to weather patterns, wind, and local temperature variations. A stationary sensor can exhibit a significant "drift" over just a few hours, as illustrated in Figure 1, where standalone readings fluctuate by several meters over a 24-hour period.
This instability hinders the implementation of reliable 3D Location-Based Services (LBS) for critical applications such as indoor navigation, emergency rescue, and precision asset tracking.
Methodology
The proposed system addresses barometric instability through three integrated mechanisms:
1. Differential Pressure Altimetry
The system utilizes a "differential" approach rather than relying on a single sensor. A fixed benchmark station at a known altitude records local pressure trends. By comparing mobile sensor data to this benchmark, large-scale atmospheric drifts are canceled out. The Laplace high-pressure equation is employed to calculate the precise height difference between the two points.
2. Temperature Difference Correction Model
Localized environmental factors, such as heating or air conditioning, can create pressure changes that mimic vertical movement. Based on the ideal gas law ($pV = nRT$), the authors derived a correction model. By measuring the rate of temperature change ($dT$), the system estimates and compensates for the pressure change ($dp$) caused solely by thermal fluctuations.
3. Human Behavioral Constraints
To filter out environmental noise, the algorithm incorporates constraints based on typical human movement. For instance, vertical movement speed is generally assumed to be less than 2 m/s (corresponding to a pressure change rate of less than 24 Pa/s).
Experimental Results
The researchers conducted real-world experiments to validate the model. A standalone pressure sensor showed a drift of approximately 2 meters over 24 hours while stationary. In contrast, the differential method maintained high stability across various height differences.
For example, in multi-point tests, a measured height difference of 1.00m was calculated as 0.9655m, representing an error of only 3.3cm. In "thermally noisy" environments simulated with electric heaters, a sensor that had not moved appeared to shift by 5.15 meters due to temperature changes alone. Applying the Temperature Difference Correction Model reduced this error to just 0.237 meters, demonstrating significant robustness.
Conclusion
This research demonstrates that differential pressure altimetry, combined with thermal correction and behavioral modeling, can achieve sub-meter vertical accuracy. While the current system requires nearby reference infrastructure, it provides a vital theoretical and technical foundation for high-precision 3D LBS. Future work will focus on integrating these models with other smartphone sensors to reduce infrastructure dependency and improve wide-area applicability.
Summary updated