Abstract
In the evolving landscape of the Internet of Things (IoT), data freshness is paramount for time-critical applications such as autonomous driving and drone-based monitoring. This paper introduces a comprehensive mathematical framework to analyze how the velocity of mobile devices—both ground-based and aerial—affects the Age of Information (AoI). By accounting for spatio-temporal correlation and base station handovers, the authors demonstrate that higher speeds generally lead to staler data, characterized by a higher Peak Age of Information (PAoI). The analysis reveals that ground users are more sensitive to movement-induced correlation than drones, providing critical insights for the design of future mobile networks.
Introduction
Data freshness is a critical requirement for modern IoT systems. Whether it is a self-driving car reporting road conditions or a drone monitoring a wildfire, delays in information can lead to catastrophic failures. Researchers use the Age of Information (AoI) metric to quantify this freshness. While most existing studies assume static IoT devices, real-world applications involve significant mobility.
Movement introduces spatio-temporal correlation: if a device fails to transmit a packet at one location, it is statistically likely to fail again nearby due to shared interference and obstacles. As a device speeds up, it traverses these interference zones more rapidly, altering the statistical reliability of its connection. This paper bridges the gap by providing a velocity-aware analysis of data freshness for moving networks.
Methodology
The authors utilize Stochastic Geometry to model the random locations of base stations (BSs) and users as Poisson Point Processes. To maintain mathematical tractability, they employ a Dominant Interferer-based Approximation, focusing on the strongest interferer while treating others as background noise.
For aerial users, the Nakagami-m Fading model is used to capture the nuances of Line-of-Sight (LoS) and Non-Line-of-Sight (NLoS) conditions. The core of the analysis involves deriving the Joint Success Probability of transmissions at different time instants, which quantifies how velocity "breaks" the correlation between successive attempts. To handle the complexity of Nakagami fading, the authors use indicator functions to replace more complex mathematical forms, enhancing computational efficiency.
Key Insights
- Velocity as a Decorrelator: Increased velocity reduces the correlation between transmission attempts. However, this decorrelation actually increases the variance and peak values of the AoI.
- Ground vs. Aerial Dynamics: Ground users exhibit higher spatio-temporal correlation than aerial users. Drones frequently transition between LoS and NLoS states, which naturally disrupts signal correlation more quickly than the relatively stable environments of ground users.
- Impact of Handover: The model explicitly accounts for handovers as devices move between base station coverage areas. The resulting "blackout" periods during these switches are major contributors to data staleness.
Results and Discussion
The proposed model's accuracy was validated through extensive Monte-Carlo Simulations, showing an excellent match, particularly in NLoS-heavy environments.
Key findings include:
- Correlation Decay: Mathematically, as velocity ($v$) approaches infinity, the correlation coefficient between transmissions drops to zero.
- PAoI Performance: The 60th percentile of PAoI increases significantly with velocity. In high-rise urban environments, aerial users experience worse PAoI than ground users due to the difficulty of maintaining stable LoS links amidst dense interference.
- Environment Sensitivity: High-density urban areas increase interference power, making velocity-aware trajectory planning essential for maintaining data freshness.
Conclusion
This study provides a robust framework for understanding the interplay between mobility and data freshness. While the current model uses a "nearest base station" handover strategy and straight-line trajectories, it sets the stage for future work. Potential extensions include modeling more complex handover triggers, realistic curving trajectories, and incorporating interference cancellation techniques to mitigate the negative effects of velocity on AoI.
Summary updated