Statistical Analysis and End-to-End Performance Evaluation of Traffic Models for Automotive Data

TL;DR

This work addresses the challenge of simulating automotive data traffic for V2X by developing statistical models of LiDAR point cloud sizes under various semantic compression levels. It introduces seven raw-data models and six representative compression configurations derived from SemanticKITTI, and validates them with a Kolmogorov-Smirnov test using Bootstrap resampling, before implementing the distributions in ns-3 for end-to-end evaluation. The results show that network metrics such as latency and throughput produced by the statistical models closely match those from real data, while enabling substantial simulation-speed gains (up to about $18\times$ faster). The study provides a practical, flexible framework for rapid, scalable evaluation of cooperative perception and V2X networks, with a quantified trade-off between compression level, data quality, and network performance.

Abstract

Autonomous driving is a major paradigm shift in transportation, with the potential to enhance safety, optimize traffic congestion, and reduce fuel consumption. Although autonomous vehicles rely on advanced sensors and on-board computing systems to navigate without human control, full awareness of the driving environment also requires a cooperative effort via Vehicle-To-Everything (V2X) communication. Specifically, vehicles send and receive sensor perceptions to/from other vehicles to extend perception beyond their own sensing range. However, transmitting large volumes of data can be challenging for current V2X communication technologies, so data compression represents a crucial solution to reduce the message size and link congestion. In this paper, we present a statistical characterization of automotive data, focusing on LiDAR sensors. Notably, we provide models for the size of both raw and compressed point clouds. The use of statistical traffic models offers several advantages compared to using real data, such as faster simulations, reduced storage requirements, and greater flexibility in the application design. Furthermore, statistical models can be used for understanding traffic patterns and analyzing statistics, which is crucial to design and optimize wireless networks. We validate our statistical models via a Kolmogorov-Smirnoff test implementing a Bootstrap Resampling scheme. Moreover, we show via ns-3 simulations that using statistical models yields comparable results in terms of latency and throughput compared to real data, which also demonstrates the accuracy of the models.

Figures (8)

• Figure 1: Overview of the system model. An autonomous vehicle generates 3D LiDAR point clouds, which may be compressed via hsc to reduce the file size. Statistical analysis via ks and Bootstrap Resampling methods is performed to identify statistical models for the size of 3D LiDAR point clouds. The accuracy of the models is assessed via statistical tests, and based on their impact on network metrics, measured via ns-3 simulations.
• Figure 2: Number of unique points $U$ and p2p-PSNR after compression vs. . Above, random point clouds from SemanticKITTI after compression for QP $\in\{8,9,10,12\}$.
• Figure 3: Average file size and encoding/decoding time vs. $\in \{0,1,\dots, 10\}$, and for $\in\{0,10,11,12,13,14\}$.
• Figure 4: Fitting plots: data histogram (light grey), empirical (dashed red), fitted pdf (light blue), fitted (orange), and QQ-plots (right).
• Figure 5: sinr and los probability (based on the model in 3gpp.38.901) vs. $d$.