Age-Aware Stochastic Hybrid Systems: Stability, Solutions, and Applications

Ali Maatouk, Mohamad Assaad, Anthony Ephremides

Introduction

In the era of real-time monitoring, industrial IoT, and vehicular networks, the freshness of data is a critical performance metric. This is formally captured by the Age of Information (AoI), defined as the time elapsed since the generation of the most recently received update. To model and analyze these dynamic systems, researchers have traditionally turned to Stochastic Hybrid Systems (SHS). SHS models combine continuous time-driven processes—such as the linear growth of age—with discrete stochastic events like packet arrivals or successful transmissions.

However, a significant limitation of standard SHS frameworks is the assumption of constant transition rates. In practical scenarios, it is often desirable for a system to be "age-aware." For instance, a sensor node holding very old data should ideally increase its transmission frequency to restore freshness. Introducing age-dependent transition rates creates a mathematical challenge: the differential equations for the $m$-th moment of age become coupled with the $(m+1)$-th moment, leading to an infinite chain of equations that cannot be solved directly.

The Age-Aware SHS Framework

This research extends the SHS framework to accommodate transition intensities that are polynomial functions of the age. This polynomial approach is versatile, covering a wide range of practical applications and allowing for the approximation of non-polynomial functions via Taylor series expansions.

A simple age-aware transition where the rate depends on the current age x(t).

The core innovation lies in addressing the "infinite chain" problem. In traditional SHS, the first moment is independent; in age-aware models, the first moment depends on the second, the second on the third, and so on. To resolve this, the authors prove the Lagrange stability and positive recurrence of these systems using Lyapunov functions, ensuring that the age does not grow to infinity and that the system reaches a steady-state distribution.

Methodology: Scaling and Moment Closure

To provide a computable solution for the infinite moment equations, the authors introduce a novel scaling technique. By tracking a scaled version of the age, $z(t) = x(t)/c$ (where $c$ is a large constant), they ensure that higher-order moments become progressively smaller. This allows for "moment closure"—truncating the system at a specific order and solving it as a finite set of linear equations. This method provides high numerical stability and accuracy compared to simple truncation, which often fails when age values are large.

Furthermore, the paper introduces a Sequential Convex Approximation (SCA) algorithm. This iterative approach allows network designers to optimize system parameters, such as back-off rates in CSMA environments, to minimize the average age. The authors prove that this algorithm converges to a stationary point, providing a practical optimization tool.

Key Results and Applications

The proposed framework was validated through Monte Carlo simulations and numerical implementations using IEEE 802.11n parameters. Key findings include:

  • High Approximation Accuracy: The moment closure method matches simulation results almost perfectly once the closure order is sufficiently high.
  • Performance Gains: In a two-user CSMA scenario, the age-aware scheme reduced the average age by up to 20% compared to the optimal age-blind configuration. This demonstrates that allowing nodes with older data to be more aggressive in channel access significantly improves overall network freshness.
  • Optimization Convergence: The SCA approach successfully calibrated parameters to reach minimum average age, grounding the theoretical work in realistic hardware constraints like $9\mu s$ slot durations.

Conclusion

By extending SHS to include age-dependent transitions and providing a robust moment closure technique, this work bridges the gap between theoretical modeling and the design of responsive, age-aware communication protocols. While currently focused on polynomial rates, future work may expand this to more complex collision models and non-linear functions, further enhancing the scalability and precision of real-time network monitoring.

Summary updated

Abstract

In this paper, we analyze status update systems modeled through the Stochastic Hybrid Systems (SHSs) tool. Contrary to previous works, we allow the system's transition dynamics to be polynomial functions of the Age of Information (AoI). This dependence allows us to encapsulate many applications and opens the door for more sophisticated systems to be studied. However, this same dependence on the AoI engenders technical and analytical difficulties that we address in this paper. Specifically, we first showcase several characteristics of the age processes modeled through the SHSs tool. Then, we provide a framework to establish the Lagrange stability and positive recurrence of these processes. Building on this, we provide an approach to compute the m-th moment of the age processes. Interestingly, this technique allows us to approximate the average age by solving a simple set of linear equations. Equipped with this approach, we also provide a sequential convex approximation method to optimize the average age by calibrating the parameters of the system. Finally, we consider an age-dependent CSMA environment where the backoff duration depends on the instantaneous age. By leveraging our analysis, we contrast its performance to the age-blind CSMA and showcase the age performance gain provided by the former.
BibTeX
@article{maatouk2021age,
  author = {Ali Maatouk and Mohamad Assaad and Anthony Ephremides},
  title = {{Age-Aware Stochastic Hybrid Systems: Stability, Solutions, and Applications}},
  journal = {IEEE J. Sel. Areas Inf. Theory},
  year = {2021},
  doi = {10.1109/jsait.2023.3337203},
}
Figure