Performance Analysis for Downlink Transmission in Multiconnectivity Cellular V2X Networks

Luofang Jiao, Jiwei Zhao, Yunting Xu, Tianqi Zhang, Haibo Zhou, Dongmei ZhaoIEEE Internet of Things Journal 2024 · DOI

TL;DR

This paper introduces a novel analytical framework for evaluating the performance of downlink transmissions in multi-connectivity Cellular V2X (C-V2X) networks. By employing stochastic geometry, the authors derive key performance metrics like coverage probability and spectral efficiency, demonstrating that multi-connectivity significantly enhances C-V2X performance, especially in high-speed scenarios.

What Problem Does This Paper Address?

The increasing number of connected vehicles in 5G and beyond (B5G) networks presents a significant challenge for reliable and high-speed C-V2X communication, particularly at high vehicle speeds. While multi-connectivity is a promising solution for improving network performance and reliability, there's a notable gap in analytical frameworks specifically for the downlink performance of multi-connectivity C-V2X. Existing research often focuses on uplink or general wireless networks, leaving the crucial downlink aspect, which supports receiving information from infrastructure, underexplored. This paper aims to fill this gap by providing a comprehensive analytical framework for downlink multi-connectivity C-V2X.

Methods

The authors propose an analytical framework based on stochastic geometry. Key methodological elements include:

  • 1-D Poisson Point Process (PPP) Modeling: Both vehicles and downlink base stations (DBSs) are modeled as 1-D PPPs on a road. This simplifies the spatial distribution analysis.
  • Joint Distance Distribution: The paper derives the joint distance distribution between a typical vehicle and its multiple connected DBSs.
  • Stochastic Geometry Tools: These tools are leveraged to derive closed-form expressions for crucial performance indicators.
  • Coverage Probability: Defined as the probability that the received Signal-to-Interference-plus-Noise Ratio (SINR) exceeds a predetermined threshold.
  • Spectral Efficiency: Defined as the amount of data transmitted per unit of bandwidth, calculated using Shannon's theory.
  • Monte Carlo Simulations: Extensive simulations are used to validate the analytical framework and demonstrate the effectiveness of multi-connectivity.

The novelty lies in applying these stochastic geometry techniques specifically to the downlink of multi-connectivity C-V2X, deriving precise analytical expressions for performance metrics, and analyzing the impact of parameters like path loss exponent and DBS density.

Datasets

Not discussed in the paper. The evaluation relies on simulated data generated through Monte Carlo simulations based on the defined system model and parameters.

Key Insights and Ideas

  • Multi-connectivity is a Game Changer for C-V2X Downlink: The core insight is that multi-connectivity significantly boosts both coverage probability and spectral efficiency in C-V2X networks, making them more suitable for demanding ITS applications.
  • Analytical Framework for Downlink: The paper provides a much-needed analytical tool for understanding and optimizing downlink performance in multi-connectivity C-V2X, moving beyond empirical or simulation-only approaches.
  • Impact of Network Parameters: The analysis reveals how path loss exponent and DBS density critically influence performance. Higher path loss exponents can improve SINR in interference-limited scenarios, while DBS density has a diminishing return effect, suggesting a trade-off between performance and cost.
  • Trade-off Between Connectivity and Cost: While increasing the number of connected DBSs improves performance, it doesn't always lead to a proportional gain and incurs higher costs, highlighting the need for balanced optimization.
  • Validation of Theory: The close agreement between analytical results and Monte Carlo simulations strongly validates the proposed framework.

Key Results

  • Coverage Probability Enhancement: Multi-connectivity (dual and triple) consistently shows higher coverage probability than single-connectivity, even when the density of multi-connectivity DBSs is lower than single-connectivity DBSs.
  • Spectral Efficiency Improvement: Dual-connectivity can improve spectral efficiency by up to 40%, and triple-connectivity by over 75% compared to single-connectivity. This gain is relatively stable with increasing path loss exponent.
  • Impact of Path Loss Exponent ($\alpha_d$): As $\alpha_d$ increases, spectral efficiency generally increases, especially in dense DBS deployments where the network is interference-limited.
  • Impact of DBS Density ($\lambda_d$): Increasing DBS density initially improves spectral efficiency, which then saturates. Multi-connectivity benefits from higher densities for a longer range before saturation.
  • Analytical vs. Simulation: The paper demonstrates strong agreement between derived analytical formulas and simulation results for both coverage probability and spectral efficiency.

Limitations and Future Work

  • Limitations:
    • The analysis is based on a simplified 1-D model, which might not fully capture the complexities of real-world 3-D urban environments.
  • The paper focuses on intra-frequency multi-connectivity. Inter-frequency or inter-RAT multi-connectivity scenarios are not explored.
  • The association policy is based on Maximum Received Power (MRP), which is a common but simplified approach.
  • Future Work:
    • Extending the analysis to 3-D environments.
  • Investigating inter-frequency and inter-RAT multi-connectivity.
  • Exploring more advanced association policies and resource allocation schemes.
  • Further research into the full potential of multi-connectivity for next-generation communication systems.

Summary updated

BibTeX
@article{jiao2024performance,
  author = {Luofang Jiao and Jiwei Zhao and Yunting Xu and Tianqi Zhang and Haibo Zhou and Dongmei Zhao},
  title = {{Performance Analysis for Downlink Transmission in Multiconnectivity Cellular V2X Networks}},
  journal = {IEEE Internet of Things Journal},
  year = {2024},
  doi = {10.1109/jiot.2023.3335233},
}
Figure