[Paper Review] Methodologies of Link-Level Simulator and System-Level Simulator for C-V2X Communication

Donglin Wang, Raja Sattiraju+3 more|arXiv (Cornell University)|Jul 9, 2019

Vehicular Ad Hoc Networks (VANETs)5 references4 citations

TL;DR

This paper presents a dual-simulator framework for C-V2X communications, combining a Link-Level (LL) simulator for physical-layer performance and a System-Level (SL) simulator for network-scale evaluation. The LL simulator generates SNR-BLER mapping curves, which are used as Link-to-System (L2S) interfaces to drive the SL simulator, enabling performance evaluation of sidelink C-V2X under varying inter-vehicle distances (IVD), velocities, and transmission frequencies, with PRR (Packet Reception Ratio) as the key performance metric. The results show that system performance improves with larger IVDs and lower vehicle speeds, while higher transmission frequencies reduce PRR due to increased load.

ABSTRACT

At the time of the development, standardization, and further improvement are vital to the modern cellular systems such as the next generation wireless communication (5G). Simulations are essential to test and optimize algorithms and procedures prior to their implementation process of the equipment manufactures. In order to evaluate system performance at different levels, accurate simulations of simple setups, as well as simulations of more complex systems via abstracted models are necessary. In this work, two new simulators for the sidelink Cooperative-Vehicle-to-Everything (C-V2X) communication have been implemented and carried out on both the physical layer (Link-Level (LL)) and network layer (System-Level (SL)). Detailed methodologies of the LL and SL simulators for C-V2X communication have been illustrated. In the LL simulator, we get the mapping curves of BLER and Signal-to-Noise-Ratio (SNR), which are used as a baseline for measuring the performance of the LL simulation. In addition, these mapping curves are used as the important Link-to-System (L2S) interfaces. The SL simulator is utilized for measuring the performance of cell networking and simulating large networks comprising of multiple eNBs and UEs. Finally, the simulation results of both simulators for CV2X communication are presented, which shows that different objectives can be met by using LL or SL simulations types.

Motivation & Objective

• To develop a standardized, modular simulation framework for C-V2X communication to support 5G network development and optimization.
• To address the limitations of real-world testing by enabling cost-effective, scalable simulation of both physical-layer and network-level performance.
• To establish a reusable Link-to-System (L2S) interface using SNR-BLER mapping for consistent performance evaluation across simulation layers.
• To evaluate the impact of key system parameters—inter-vehicle distance (IVD), mobile velocity, and transmission frequency—on C-V2X system performance.

Proposed method

• The LL simulator models a single-link communication using the Extended Vehicular A (EVA) channel model, simulating Rayleigh fading and Doppler effects for vehicular environments.
• It generates a mapping table between Signal-to-Noise Ratio (SNR) and Block Error Rate (BLER) to serve as the core L2S interface for system-level evaluation.
• The SL simulator uses the LL-derived SNR-BLER mapping to compute Packet Reception Ratio (PRR) across multiple UEs in a networked scenario with multiple eNBs and UEs.
• A probabilistic algorithm is implemented in the SL simulator: for each receiver, a random value is compared against the BLER to determine successful reception, enabling statistical PRR estimation.
• The SL simulator evaluates performance under different IVDs, mobile velocities (up to 500 km/h), and transmission frequencies (10 Hz and 20 Hz), with PRR as the primary performance metric.
• The simulation framework is validated in a highway scenario with network-assisted DC-V2X using 3GPP Release-14 sidelink (PC5) mode 3, where resource allocation is managed by the base station.

Experimental results

Research questions

• RQ1 How does inter-vehicle distance (IVD) affect the Packet Reception Ratio (PRR) in sidelink C-V2X communications under varying mobility conditions?
• RQ2 What is the impact of mobile velocity on system performance, particularly due to the Doppler effect, in a C-V2X network scenario?
• RQ3 How does transmission frequency influence system load and PRR in a C-V2X sidelink communication setup?
• RQ4 To what extent can the LL simulator’s SNR-BLER mapping be effectively reused in the SL simulator to ensure consistency and accuracy in system-level performance evaluation?
• RQ5 What are the trade-offs between system load (number of UEs) and performance (PRR) in a high-mobility, multi-UE C-V2X environment?

Key findings

• PRR increases significantly with larger inter-vehicle distances (IVD), rising from 31.56% at 10 m to 96.49% at 100 m when all UEs move at 100 km/h.
• At a fixed IVD of 100 m, PRR drops from 94.50% at 100 km/h to 78.72% at 500 km/h due to increased Doppler spread and channel variation.
• Reducing transmission frequency from 20 Hz to 10 Hz improves PRR, with a 10 m IVD increasing PRR from 15.67% to 31.56%, as lower frequency reduces system load.
• The SL simulator successfully leverages the LL simulator’s SNR-BLER mapping to enable accurate, scalable system-level performance evaluation across diverse mobility and deployment scenarios.
• System performance is degraded by high vehicle density and high mobility, indicating that load control and mobility-aware resource allocation are critical for reliable C-V2X communication.