Performance Analysis of Age of Information in Ultra-Dense Internet of Things (IoT) Systems With Noisy Channels

Bo Zhou, Walid Saad

TL;DR

This paper analyzes the "Age of Information" (AoI) in ultra-dense IoT systems with noisy wireless channels, considering different device behaviors (policies) and the impact of transmission feedback. They develop analytical models and a mean-field approximation to show that preemption and specific device policies can significantly reduce AoI, and that AoI doesn't always improve with higher data arrival rates.

What Problem Does This Paper Address?

The Internet of Things (IoT) is rapidly expanding, with massive numbers of devices needing to send real-time status updates to receivers. The "Age of Information" (AoI) is a crucial metric for measuring how fresh this information is. In ultra-dense IoT systems, devices often share wireless channels using Carrier Sense Multiple Access (CSMA). However, existing research often assumes perfect transmission or doesn't fully account for the complexities of noisy channels and the decision-making processes of devices after a transmission attempt. This paper addresses the challenge of analyzing AoI in such systems, specifically considering:

  • Noisy Channels: Transmissions can fail.
  • Transmission Feedback: Devices may or may not know if their transmission was successful.
  • Device Policies: How devices react to transmission outcomes (e.g., retransmit, wait, or give up).
  • Preemption: Whether a new incoming packet can interrupt an ongoing transmission.

Understanding how these factors influence AoI is critical for designing efficient and responsive IoT systems.

Methods

The authors employ a combination of analytical and approximation techniques:

  • Stochastic Hybrid Systems (SHS): This is the core modeling technique. SHS allows for the analysis of systems with both discrete state transitions (like a device changing its status: Idle, Waiting, Service) and continuous-time evolution (like the age of information itself).
  • Continuous-Time Markov Chains (CTMC): The discrete states of the devices are modeled as CTMCs.
  • Closed-Form Expressions: For smaller systems, the authors derive exact mathematical formulas for the average AoI under different policies and scenarios (with/without preemption, with/without feedback).
  • Mean-Field Approximation: For large numbers of devices (approaching infinity), the complex interactions are simplified by assuming each device's behavior is influenced by the average behavior of all other devices. This leads to a deterministic dynamic system (an Ordinary Differential Equation - ODE) that approximates the system's behavior.
  • Simulation: Numerical simulations are used to validate the analytical results and the mean-field approximation, especially for practical system sizes.

Datasets

Not discussed in the paper. The paper focuses on theoretical modeling and simulation rather than empirical data from real-world deployments.

Key Insights and Ideas

  • Preemption is Key: Schemes with preemption in service consistently achieve lower average AoI than those without, regardless of the device policy. This means interrupting an old transmission to send a newer one is beneficial for information freshness.
  • Policy Matters: Different device policies (how they react to transmission outcomes) have a significant impact on AoI. Policy (S), where a device stays in service until successful transmission, generally yields the lowest AoI.
  • Feedback is Crucial (but complex): The presence or absence of feedback fundamentally changes device behavior and AoI.
  • AoI Doesn't Always Improve with Arrival Rate: Counter-intuitively, increasing the arrival rate (λ) doesn't always decrease AoI. This is because higher arrival rates can lead to more devices entering the "waiting" or "service" states, potentially increasing congestion and delays.
  • Mean-Field Approximation is Accurate: The developed mean-field approach provides accurate predictions of AoI even for systems with a relatively small number of devices, making it a powerful tool for analyzing large-scale IoT deployments.
  • Parameter Sensitivity: AoI is sensitive to various system parameters. Better channel conditions (higher success probability 'p', higher service rate 'μ') and more communication resources (lower ratio 'γ') generally lead to lower AoI.

Key Results

  • Closed-form expressions for average AoI are derived for three device policies (I, W, S) under schemes with and without preemption, and with and without feedback.
  • Scheme with preemption always outperforms the scheme without preemption for all policies.
  • Policy (S) achieves the smallest average AoI compared to policies (I) and (W).
  • Policy (W) outperforms policy (I) only when channel utilization is low.
  • The mean-field approximation is shown to be accurate for analyzing large-scale systems, with simulation results closely matching analytical predictions.
  • Parameter analysis shows that AoI decreases with increasing service rate (μ), waiting rate (w), and successful transmission probability (p), and increases with the ratio (γ).

Limitations and Future Work

  • Idealized CSMA Assumptions: The paper relies on idealized CSMA assumptions (instantaneous sensing, no hidden nodes), which might not hold in all real-world scenarios.
  • Focus on Average AoI: While average AoI is a key metric, other metrics like peak AoI or AoI variance might be important for certain applications.
  • Future Work: The authors suggest investigating the optimization of system parameters to minimize AoI under constraints such as energy cost or throughput. They also mention extending the analysis to other network performance metrics.

Summary updated

Abstract

In this paper, a dense Internet of Things (IoT) monitoring system is studied in which a large number of devices contend for transmitting timely status packets to their corresponding receivers over wireless noisy channels, using a carrier sense multiple access (CSMA) scheme. When each device completes one transmission, due to possible transmission failure, two cases with and without transmission feedback to each device must be considered. Particularly, for the case with no feedback, the device uses policy (I): It will go to an idle state and release the channel regardless of the outcome of the transmission. For the case with perfect feedback, if the transmission succeeds, the device will go to an idle state, otherwise it uses either policy (W), i.e., it will go to a waiting state and re-contend for channel access; or it uses policy (S), i.e., it will stay at a service state and occupy this channel to attempt another transmission. For those three policies, the closed-form expressions of the average age of information (AoI) of each device are characterized under schemes with and without preemption in service. It is shown that, for each policy, the scheme with preemption in service always achieves a smaller average AoI, compared with the scheme without preemption. Then, a mean-field approximation approach with guaranteed accuracy is developed to analyze the asymptotic performance for the considered system with an infinite number of devices and the effects of the system parameters on the average AoI are characterized. Simulation results show that the proposed mean-field approximation is accurate even for a small number of devices. The results also show that olicy (S) achieves the smallest average AoI compared with policies (I) and (W), and the average AoI does not always decrease with the arrival rate for all three policies.
BibTeX
@article{zhou2021performance,
  author = {Bo Zhou and Walid Saad},
  title = {{Performance Analysis of Age of Information in Ultra-Dense Internet of Things (IoT) Systems With Noisy Channels}},
  journal = {IEEE Transactions on Wireless Communications},
  year = {2021},
  doi = {10.1109/twc.2021.3122841},
}
Figure