On the Analysis of Platooned Vehicular Networks on Highways

Technical Summary: On the Analysis of Platooned Vehicular Networks on Highways

Problem Statement
Vehicular platooning, the coordinated movement of vehicles in close proximity, offers significant benefits regarding fuel efficiency, emissions, and traffic flow. However, the communication dynamics required to maintain these platoons on highways remain underexplored. While platooning facilitates Vehicle-to-Vehicle (V2V) communication, the close clustering of vehicles may alter the load distribution on Road-Side Units (RSUs) used for Vehicle-to-Infrastructure (V2I) connectivity. Specifically, the correlation of vehicle locations in a platoon could lead to "bursty" traffic loads, potentially causing some RSUs to become overloaded while others remain underutilized. This paper addresses the lack of comprehensive analysis regarding how platooning affects RSU load distribution, coverage probability (CP), rate coverage (RC), and the reliability of these metrics across different spatial distributions of vehicles and infrastructure.

Methodology
The authors employ a stochastic geometry framework to model vehicular networks on a highway. The system consists of RSUs and Vehicular Users (VUs) distributed along a one-dimensional line.

• Infrastructure Model: RSUs are modeled as a homogeneous 1D Poisson Point Process (PPP) with density $\lambda_r$.
• Traffic Scenarios: Two distinct traffic models are analyzed:
  1. Non-Platooned Traffic Scenario (N-PTS): VUs move independently and are modeled as a 1D PPP with density $\lambda$.
  2. Platooned Traffic Scenario (PTS): VUs are clustered. This is modeled using a 1D Matérn Cluster Process (MCP), where parent points (platoon leaders) follow a PPP, and daughter points (platoon members) are distributed within a ball of radius $a$ around the parent.
• Performance Metrics: The study derives closed-form expressions and distributions for:
  • Load Distribution: The number of VUs served by a typical RSU and a tagged RSU (serving a specific typical vehicle).
  • Connectivity: The "connectivity degree" for V2V communication and the Coverage Probability (CP) for V2I communication.
  • Reliability: The Meta Distribution (MD) of the Signal-to-Interference-plus-Noise Ratio (SINR) and achievable rates, which quantifies the variability of link success probabilities across the network.
• Analytical Tools: The analysis utilizes Probability Generating Functions (PGFs), Moment Generating Functions (MGFs), and properties of the 1D MCP and PPP to derive exact expressions for mean, variance, skewness, and higher-order moments.

Key Contributions
The paper provides the following specific analytical contributions:

1. Load Distribution Derivation: Closed-form expressions for the load distribution on RSUs under both PTS and N-PTS. The authors highlight that the difference in load arises from the spatial correlation of VU locations in platoons. They further derive the mean, variance, and skewness of the load to characterize the burstiness of traffic.
2. Active RSU Density: Using the load distributions, the paper derives the probability of an RSU being active (serving at least one vehicle) and the resulting density of active RSUs, which is critical for interference modeling.
3. V2V and V2I Performance Analysis:
   • V2V: The study analyzes the connectivity degree (number of neighbors within range) for typical VUs in both scenarios.
   • V2I: The authors derive the Coverage Probability (CP) and Rate Coverage (RC) for the typical VU. Crucially, they derive the exact Meta Distribution (MD) for SINR and rate, providing a deeper understanding of link reliability beyond average metrics.
4. Numerical Validation: Theoretical findings are validated through simulations, demonstrating the impact of VU density and the extent of platooning on network performance.

Results
Numerical results indicate that the Platooned Traffic Scenario (PTS) significantly alters network dynamics compared to independent movement:

• Load Burstiness: PTS leads to increased burstiness in RSU load distribution due to the clustering of vehicles.
• RSU Efficiency: Contrary to the intuition that clustering might always overload infrastructure, the analysis shows that PTS can improve RSU efficiency. This is attributed to the specific nature of the load distribution where higher VU density under platooning conditions results in more efficient utilization of active RSUs compared to the N-PTS baseline.
• Reliability: The Meta Distribution analysis reveals how VU density and platooning affect the variability of link success probabilities, offering insights into the reliability of connections for individual vehicles rather than just network averages.

Significance and Claims
The paper claims to fill a gap in the literature by providing a tractable stochastic geometry framework specifically for analyzing highway platooning scenarios. While previous works have analyzed load distribution in two-dimensional networks or focused on connectivity without considering the specific load dynamics of platoons, this work offers a comprehensive analysis of per-RSU load, SINR, and rate distributions along with their meta distributions for one-dimensional highway settings. The authors assert that understanding these load dynamics is critical for maintaining stable platoon operations and optimizing infrastructure resource allocation. The study does not propose new hardware or control algorithms but rather provides the analytical foundation necessary to evaluate the performance of existing communication technologies (V2V and V2I) under the specific spatial constraints of vehicular platooning.