A QoS Based Adaptive Backoff Scheme for Vehicular Ad Hoc Networks

Tianjiao Zhang, Qi Zhu2018 · DOI

TL;DR

This paper introduces a novel adaptive backoff scheme for vehicular ad hoc networks (VANETs) that prioritizes data based on its delay requirements. By dynamically adjusting backoff parameters, the scheme significantly reduces average data delay and ensures that different data types meet their specific Quality of Service (QoS) needs, especially in crowded network conditions.

What Problem Does This Paper Address?

Vehicular ad hoc networks (VANETs) are crucial for intelligent transportation systems, enabling vehicle-to-vehicle and vehicle-to-infrastructure communication. A key challenge in VANETs is the Media Access Control (MAC) protocol, which governs how nodes share the communication channel. Existing MAC protocols, like IEEE 802.11p, often use a single backoff process that struggles to handle diverse data types with varying delay sensitivities. Emergency data, for instance, requires extremely low latency, while other data types might be more tolerant. Furthermore, the dynamic nature of VANETs, with constantly changing numbers of vehicles, exacerbates this problem, leading to increased delays and potential safety issues. Existing backoff schemes are not flexible enough to adapt to these varying conditions and multi-class traffic demands, leading to suboptimal network performance.

Methods

The paper proposes an Adaptive Backoff Scheme that enhances existing algorithms like Binary Exponential Backoff (BEB) and Quadratic Backoff (QB). The core idea is to classify data into different priority levels based on their delay requirements. For each data type, a distinct backoff strategy is employed, and crucial backoff parameters, specifically the probability of resetting the backoff stage ($\beta_k$), are dynamically adjusted.

Key methodological contributions include:

  • Priority-Based Data Classification: Data is categorized based on its delay tolerance, with emergency data receiving the highest priority.
  • Enhanced Backoff Algorithms (eBEB and eQB): The proposed scheme modifies BEB and QB by introducing adaptive $\beta_k$ values.
  • Two-Dimensional Markov Chain Analysis: A detailed mathematical model is developed to analyze the performance (throughput and delay) of the proposed backoff scheme under various conditions.
  • Particle Swarm Optimization (PSO): To find the optimal $\beta_k$ values that minimize total delay while respecting individual data type delay constraints, a PSO algorithm is employed. This is crucial because the delay optimization problem is complex and non-linear.

Datasets

Not discussed in the paper. The evaluation is based on simulations using parameters representative of a highway scene and system channel characteristics.

Key Insights and Ideas

  • Dynamic Adaptation is Key: The effectiveness of VANET MAC protocols hinges on their ability to adapt to dynamic network conditions and diverse traffic demands. A static approach is insufficient.
  • Priority Matters: Differentiating traffic based on delay requirements is essential for ensuring critical data (like emergency alerts) gets through with minimal latency.
  • $\beta_k$ is a Powerful Control Knob: The probability of resetting the backoff stage ($\beta_k$) offers a flexible way to manage contention and prioritize traffic without drastically altering fundamental MAC parameters like contention window sizes.
  • Crowded Networks Highlight Limitations: The limitations of traditional backoff schemes become most apparent in high-density scenarios, where the proposed adaptive approach shows its greatest benefit.
  • eBEB Outperforms eQB in this Context: While QB generally offers better performance than BEB, the adaptive scheme makes the enhanced BEB (eBEB) perform better than the enhanced QB (eQB) in the simulated scenarios, suggesting its suitability for certain VANET traffic patterns.

Key Results

  • Reduced Average Total Delay: The proposed adaptive backoff scheme significantly reduces the average total delay, especially in crowded network conditions (e.g., when the number of vehicles exceeds 50). For instance, in some scenarios, the delay reduction was nearly 15% compared to existing methods.
  • Adaptive Delay Control: The scheme effectively regulates the delay for each data type, ensuring that higher priority data meets its stringent requirements while lower priority data also performs acceptably.
  • Improved Throughput: While the primary focus is delay reduction, the adaptive scheme also maintains high throughput, achieving over 90% of the theoretical saturated throughput even in challenging network conditions.
  • Optimal $\beta_k$ Values: The study demonstrates that $\beta_1$ (for higher priority data) is consistently larger than $\beta_2$ (for lower priority data), reflecting the need for more frequent resets for critical traffic.

Limitations and Future Work

  • Simulation-Based Evaluation: The results are based on simulations, and real-world testing would be necessary to fully validate the scheme's performance.
  • Specific Scenario: The study focuses on a highway scenario. Its applicability to other VANET environments (e.g., urban intersections) might require further investigation.
  • Parameter Tuning: While PSO is used to optimize $\beta_k$, the performance might still be sensitive to the initial parameter settings of the PSO algorithm and the underlying MAC protocol.
  • Future Work: The authors suggest exploring the application of this adaptive backoff scheme to other MAC protocols and investigating its performance in more complex and diverse vehicular network scenarios. Further research could also focus on more sophisticated methods for data classification and dynamic adjustment of other backoff parameters.

Summary updated

BibTeX
@misc{zhang2018qos,
  author = {Tianjiao Zhang and Qi Zhu},
  title = {{A QoS Based Adaptive Backoff Scheme for Vehicular Ad Hoc Networks}},
  year = {2018},
  doi = {10.3390/s18124421},
}
Figure