[Paper Review] Technical Report - MillimeterWave Communication in Vehicular Networks: Coverage and Connectivity Analysis

Marco Giordani, Andréa Zanella+1 more|arXiv (Cornell University)|Apr 26, 2017

Millimeter-Wave Propagation and Modeling3 references3 citations

TL;DR

This technical report presents a mathematical model for analyzing coverage and connectivity in millimeter wave vehicular networks using a Poisson point process to model infrastructure nodes and a realistic 28 GHz channel model from New York City measurements. Key findings show that throughput initially increases with node density due to improved signal strength but degrades at high densities due to interference and beam misalignment, while lower vehicle speeds and shorter beam update intervals improve communication duration and throughput.

ABSTRACT

In this technical report (TR), we describe the mathematical model we developed to carry out a preliminary coverage and connectivity analysis in an automotive communication scenario based on mmWave links. The purpose is to exemplify some of the complex and interesting tradeoffs that have to be considered when designing solutions for mmWave automotive scenarios.

Motivation & Objective

• To model and analyze coverage and connectivity in mmWave-based vehicular networks under realistic propagation conditions.
• To investigate the tradeoffs between node density, vehicle speed, beam update intervals, and MIMO configuration on network performance.
• To quantify how beamforming gain, path loss, shadowing, and mobility affect communication reliability and throughput in mmWave V2X scenarios.
• To evaluate the impact of interference and beam misalignment on connectivity duration and data rates in dense urban deployments.
• To provide design insights for beam tracking protocols tailored to high-mobility vehicular environments with directional mmWave links.

Proposed method

• Models infrastructure nodes as a Poisson Point Process (PPP) with density ρ nodes/km to represent random deployment of mmWave base stations.
• Employs a realistic 28 GHz mmWave channel model based on real-world measurements in New York City, including LoS, NLoS, and outage states with distance-dependent probabilities.
• Uses a pathloss model with parameters α and β derived from measurements, and incorporates shadowing ξ ∼ N(0, σ²) with σ² from empirical data.
• Models the time-varying channel matrix H(t,f) using K clusters (Poisson-distributed) and Lk subpaths per cluster, with small-scale fading based on Doppler shift and delay spread.
• Computes beamforming gain G_BF using matched filtering between transmit and receive beamforming vectors, maximizing signal power via beam alignment.
• Derives average throughput as B = E[R(d)] × E[T_comm]/T_RTO, where E[T_comm] is the expected communication time per slot and E[R(d)] depends on mean distance 1/ρ.

Experimental results

Research questions

• RQ1 How does increasing the density of infrastructure nodes (ρ) affect the average throughput and connectivity duration in mmWave V2I networks?
• RQ2 What is the impact of vehicle speed (V) on beam alignment stability and communication time per slot (E[T_comm])?
• RQ3 How do beam update intervals (T_RTO) influence the average throughput, and what are the trade-offs with signaling overhead?
• RQ4 How does MIMO configuration (e.g., 64×16 vs. 4×4) affect the achievable communication range and throughput?
• RQ5 At what node density does interference begin to dominate over path loss reduction, leading to throughput degradation?

Key findings

• Throughput initially increases with node density ρ due to reduced mean distance and improved signal-to-interference-plus-noise ratio (SINR), peaking around ρ ≈ 40 nodes/km.
• Beyond ρ ≈ 40 nodes/km, throughput decreases due to increased interference from nearby base stations and reduced beam alignment robustness from beam narrowing at short distances.
• Lower vehicle speeds (e.g., 10 km/h vs. 130 km/h) significantly increase E[T_comm]/T_RTO, improving average throughput by reducing the likelihood of losing connectivity during a slot.
• Shorter beam update intervals (T_RTO) improve throughput by reducing disconnection time, but the benefit is limited by signaling overhead not accounted for in the model.
• Larger MIMO arrays (e.g., 64×16) increase the achievable communication range and throughput compared to smaller configurations (e.g., 4×4), due to higher beamforming gain.
• The optimal system design balances node density, beam update frequency, and MIMO size to maximize throughput while minimizing beam misalignment and interference.