PaperlayerView exact revision on arXiv

Source-linked AI summary

Realistic Characterization and Modeling of Vehicle-Mounted Antenna Radiation Patterns Based on Large-Scale Full-Vehicle Measurements

Guotian Ji, Tianhao Wang, Changhao Xi, Guokai Jiang, Jiaxu Feng, Zhiqiang Yuan, Wei Fan

arXiv:2609.36643v1eess.SP

TL;DR

Large-scale, measurement-driven evidence on how vehicle integration changes antenna patterns across platforms and installations remains limited. This paper measures complete vehicles, characterizes the resulting patterns, and proposes a statistical model; the measurements show substantial variation by vehicle structure, antenna location, and communication function.

  • Problem

    Large-scale empirical studies remain scarce, leaving vehicle-induced pattern distortion across platforms, mounting locations, frequencies, and wireless functions insufficiently characterized.

    Source: §I · p. 2

  • Method

    The paper establishes a unified full-vehicle spherical near-field dataset across ten commercial platforms and proposes a statistical model using physically interpretable parameters to generate realistic installed patterns.

    Source: §I · p. 2, A. Measurement Platform Overview · p. 3

  • Results

    Measured patterns vary substantially across vehicle structures, antenna locations, and communication functions, while the proposed model captures dominant characteristics of measured patterns.

    Source: §I · p. 2, §V · p. 10

  • Takeaways & Limitations

    The dataset, characterization, and modeling approach provide a foundation for realistic antenna evaluation and wireless-system simulations in vehicular communication applications.

    Source: C. Obtained Measurement Dataset · p. 4, C. Vehicle-Mounted Antenna Modelling · p. 9

Abstract

from arXiv · show

Vehicle-mounted antenna radiation patterns can be significantly altered after installation due to interactions with the vehicle body. However, existing studies mainly focus on simulations or isolated antenna characterization, while large-scale experimental investigations remain limited. This paper presents a comprehensive experimental analysis of vehicle-mounted antenna radiation patterns based on full-vehicle spherical near-field measurements. A large-scale dataset covering ten commercial vehicle platforms, multiple antenna mounting locations, and various wireless functions, including 4G, 5G, V2X, and GNSS, is established under a unified measurement framework. Based on the measured results, representative case studies and statistical characterization are performed to investigate the impacts of vehicle installation on antenna radiation behavior. The results reveal substantial variations across mounting locations and wireless functions, highlighting the strong influence of vehicle integration on realistic antenna performance. Furthermore, a novel statistical model is proposed to generate realistic vehicle-mounted antenna patterns using a small set of physically interpretable parameters. The obtained dataset, characterization results, and modeling approach provide a foundation for realistic antenna evaluation and wireless system simulations in vehicular communication applications.

I. INTRODUCTION

Vehicle integration can substantially reshape antenna radiation, but systematic, measurement-driven evidence across vehicle platforms, mounting locations, frequencies, and wireless functions remains scarce. This paper addresses that gap with full-vehicle measurements, empirical characterization, and a statistical model for realistic installed patterns.

Motivation

  • Installed radiation behavior matters because it helps determine communication reliability, coverage continuity, and link robustness.The introduction motivates realistic vehicle-mounted antenna characterization in the context of growing demand for reliable vehicular wireless communication.Source: §I · p. 1
  • Vehicle bodies introduce scattering, diffraction, reflection, and shadowing that can reshape installed patterns, causing directional imbalance, radiation nulls, and coverage degradation.The paper argues the vehicle body should be treated as part of the radiating system rather than merely a support structure.Source: §I · p. 1

Research gap

  • Large-scale empirical studies of vehicle-induced pattern distortion remain scarce, leaving cross-platform, location, frequency, and function characteristics insufficiently understood.Existing work has mainly examined simulations, platform validation, virtual drive testing, compact modeling, or single cases.Source: §I · pp. 1–2

Contributions

  • The study conducts a unified high-resolution spherical near-field measurement campaign spanning ten commercial vehicle types, multiple mounting locations, wireless functions, and frequency bands.The study also characterizes effects of vehicle structure, mounting position, operating frequency, and antenna function, then proposes a statistical model with physically interpretable parameters.Source: §I · p. 2

II. FULL-VEHICLE MEASUREMENT CAMPAIGN

The campaign uses a high-precision, single-probe spherical near-field platform and measures complete vehicles under a unified, repeatable framework. Its dense angular sampling and consistent calibration and processing are designed to capture installed antenna behavior across vehicle types and deployment settings.

A. Measurement Platform Overview

  • The measurement facility uses a 6 m test distance and a 6 m-radius scanning arm, with an elevation rotation range of 0°–105°.The fully anechoic chamber accommodates vehicles up to 6 m in length and operates up to 40 GHz.Source: A. Measurement Platform Overview · p. 3
  • The single-probe mechanical scanner supports high-resolution characterization from 400 MHz to 8 GHz, with 0.05° positioning accuracy to capture pattern ripples and sharp vehicle-induced nulls.The range covers 5G NR n41/n78/n79, V2X, and GNSS; the single-probe architecture avoids mutual coupling between sensors.Source: A. Measurement Platform Overview · p. 3

B. Measurement Campaign

  • Measurements cover complete vehicles—not isolated antennas—across ten commercial platforms, multiple mounting locations, and 4G, 5G, V2X, and GNSS systems.The dataset includes sedans, SUVs, and MPVs, with rooftop, mirror, bumper, windshield, and interior placements, plus single- and multiport configurations.Source: B. Measurement Campaign · p. 3
  • A shared chamber, hardware, sampling configuration, coordinate system, calibration, and post-processing procedure provide a unified basis for cross-platform comparisons.Patterns span full azimuth and the upper hemisphere, with angular resolution of 0.5° or 1° depending on frequency and scenario, followed by near-field-to-far-field transformation.Source: B. Measurement Campaign · p. 3

C. Obtained Measurement Dataset

The campaign produces a broad dataset of full-vehicle antenna patterns across commercial platforms, installations, and wireless functions. Measured examples and cross-scenario comparisons show substantial variation in installed patterns, establishing an empirical basis for further characterization and modeling.

Obtained Measurement Dataset

  • The dataset spans rooftop, bumper, mirror, windshield, and interior antennas for 4G, 5G, V2X, and GNSS applications.It covers ten commercial vehicle platforms and multiple operating bands.Source: C. Obtained Measurement Dataset · pp. 3–4
  • Measured patterns differ substantially across vehicle structures, antenna locations, and communication functions, even within the same frequency band.Representative patterns show directional imbalance, localized scattering lobes, and deep nulls across installation scenarios.Source: C. Obtained Measurement Dataset · p. 4
  • The resulting dataset provides an empirical foundation for studying vehicle-induced pattern distortion, statistical radiation characteristics, and realistic installed antenna behavior.It also supports subsequent analysis, modeling, and system-level evaluation for vehicular communication systems.Source: C. Obtained Measurement Dataset · p. 4

III. CHARACTERIZATION METRICS FOR VEHICLE-MOUNTED ANTENNA PATTERN ANALYSIS

The section introduces metrics for characterizing installed-pattern distortion, especially horizontal omnidirectionality, across vehicle platforms, locations, frequencies, and wireless functions.

III. CHARACTERIZATION METRICS FOR VEHICLE-MOUNTED ANTENNA PATTERN ANALYSIS

  • The analysis uses σH, HPR, PAG, and UHE to characterize pattern distortion across vehicle platforms, locations, frequencies, and wireless functions.The metrics address omnidirectional uniformity, directional gain, and service-region coverage.Source: §III · p. 4

A. Horizontal Omnidirectionality Metrics

B. Partial Average Gain

Partial Average Gain characterizes aggregated radiation capability over a specified service angular region.

B. Partial Average Gain

  • PAG evaluates aggregated radiation capability over a specified service angular region using a weighted spatial average.The region is bounded by θA and θB in elevation and ϕA and ϕB in azimuth, and divided into p and q intervals.Source: B. Partial Average Gain · p. 7
  • PAG is one of four metrics used to statistically characterize the complete measured dataset.The other metrics are σH, HPR, and UHE.Source: B. Partial Average Gain · p. 7

A. Specific Case Studies

For the same function on Vehicle Type 2, glass-integrated and roof-mounted sharkfin antennas show different pattern shapes and statistical metrics associated with their installation environments.

1) The Effect of Installation Locations:

  • The glass-integrated antenna has a more irregular pattern, with stronger local fluctuations, forward-biased main radiation, and partial rear-side shadowing.The inclined glass surface and nearby vehicle-body blockage are cited as contributing factors.Source: 1) The Effect of Installation Locations · p. 7
  • The sharkfin antenna has a smoother, more omnidirectional horizontal pattern, attributed to the more open radiation environment above the roof plane.Its σH and HPR are slightly lower than those of the glass-integrated antenna.Source: 1) The Effect of Installation Locations · p. 7
  • The azimuthally summed component cut combines orthogonal polarization gains and weights them by sin θ to account for the spherical surface element.Source: 1) The Effect of Installation Locations · p. 7

C. Upper-Hemisphere Efficiency (UHE)

Vehicle installation can intensify pattern distortion at a more obstructed mounting location, while UHE evaluates normalized radiation capability in the upper hemisphere. The section also examines selection-combined patterns from distributed antennas using UHE and horizontal omnidirectionality.

C. Upper-Hemisphere Efficiency (UHE)

  • At Vehicle Type 10’s right-rear bumper, σH and HPR rise from 6.12 dB and 20.22 dB to 9.20 dB and 28.67 dB, while PAG falls from 9.72 dB to 6.07 dB.The comparison attributes the stronger distortion to cabin shadowing, rear-body blockage, and local scattering.Source: C. Upper-Hemisphere Efficiency (UHE) · p. 7
  • UHE measures normalized radiation capability within the upper hemisphere, especially relevant to GNSS reception and links to cellular base stations.The definition normalizes integrated linear realized gain over the full solid angle by the isotropic-radiator value.Source: C. Upper-Hemisphere Efficiency (UHE) · p. 7

2) Integrated Pattern Analysis:

  • The section investigates selection-combined V2X patterns from distributed antennas for UHE and horizontal-omnidirectionality analysis.The combined pattern selects the maximum gain of the two antennas at each angular sample; the isotropic-radiator normalization uses 1/(4π).Source: 2) Integrated Pattern Analysis · p. 7

IV. RESULTS AND MODELING ANALYSIS

The results examine installation-location and function-specific differences in measured vehicle-mounted antenna patterns. Distributed antennas are also evaluated through combined patterns, while GNSS, V2X, and 5G patterns reflect distinct radiation requirements.

IV. RESULTS AND MODELING ANALYSIS

  • At 5915 MHz, combining two antennas by selecting maximum gain at each angular sample produces fewer deep nulls and more uniform azimuthal coverage than the individual patterns.The effect is particularly evident for Vehicle Type 10, where one bumper antenna’s blockage is compensated by the antenna mounted in another direction.Source: §IV · pp. 7–8, Figure 5 · p. 8
  • Vehicle Type 2’s windshield-integrated antenna has a more irregular pattern with stronger local fluctuations than its roof-mounted sharkfin antenna.The comparison attributes the windshield pattern to the inclined glass surface and nearby vehicle-body blockage.Source: Figure 4 · p. 8

3) Function-Specific Radiation Analysis:

  • GNSS achieves 30.23% UHE, while V2X and 5G have HPR values of 15.22 dB and 19.23 dB, respectively.GNSS concentrates radiation toward the upper hemisphere; V2X and 5G emphasize horizontal coverage and are more sensitive to azimuthal nulls and coverage discontinuities.Source: 3) Function-Specific Radiation Analysis · p. 8, Figure 6 · p. 8
  • No single metric adequately characterizes all antenna functions; σH and HPR should be considered with service-region metrics such as PAG and UHE.The measured GNSS, V2X, and 5G patterns differ in their upper-hemisphere and near-horizon coverage emphasis.Source: 3) Function-Specific Radiation Analysis · p. 8

B. Statistical Characterization

Statistical characterization groups measurements by wireless function and treats frequency points as independent samples within each group. The CDFs show broadly similar mean horizontal gain fluctuation, but differences in azimuthal coverage, service-region gain, and upper-hemisphere efficiency.

B. Statistical Characterization

  • Across 4G/5G, GNSS, and V2X/C-V2X antennas, σH means were 4.98, 4.53, and 4.33; their largely overlapping distributions indicate broadly similar average horizontal-gain fluctuation.HPR means were 10.50, 14.01, and 13.93, with variances of 11.18, 7.86, and 9.98, respectively.Source: B. Statistical Characterization · p. 9
  • GNSS has mean PAG of 0.71 dBi, compared with -10.08 dBi for 4G/5G and -6.30 dBi for V2X/C-V2X.The broad distributions across all functions indicate variation in service-region coverage performance.Source: B. Statistical Characterization · p. 9, Figure 8 · p. 9
  • Mean UHE is 32.39% for GNSS, 28.14% for 4G/5G, and 17.89% for V2X/C-V2X.The 4G/5G distribution has significantly larger spread than GNSS, indicating stronger dependence on installation conditions.Source: B. Statistical Characterization · p. 9, Figure 9 · p. 9

C. Vehicle-Mounted Antenna Modelling

The model statistically generates realistic vehicle-mounted patterns rather than reproducing individual measurements, using physically interpretable parameters to represent vehicle-induced effects. A V2X example shows agreement with the measured radiation trend, while extracted parameters distinguish mounting-position patterns.

C. Vehicle-Mounted Antenna Modelling

  • The model targets efficient system-level simulation and statistically generates realistic patterns instead of reproducing individual measurements.Its stochastic residual R(ϕ) is a zero-mean Gaussian random Fourier series accounting for small-scale angular ripples from local scattering and irregular vehicle-body interactions.Source: C. Vehicle-Mounted Antenna Modelling · pp. 9–10
  • The model uses G0, Q, and ϕB, extracted from measured patterns, to characterize baseline gain, distortion severity, and dominant blockage direction.The parameters are summarized in Table III.Source: C. Vehicle-Mounted Antenna Modelling · p. 9, Table III · p. 10
  • The model combines a directional-bias cosine, a vehicle-body shadowing notch, and stochastic angular ripples to represent large- and small-scale pattern effects.G0, Q, and ϕB represent baseline radiation level, distortion strength, and dominant blockage direction; α balances bias and suppression, while σB controls notch width.Source: C. Vehicle-Mounted Antenna Modelling · p. 10
  • For a Vehicle Type 10 left-front bumper V2X antenna at 5915 MHz, generated patterns reproduced observed angular fluctuations and agreed with the measured radiation trend.The measured-pattern parameters were G0 = 3.0 dB, Q = 12.2 dB, and ϕB = 229.5°; Fig. 10 compares the measurement with model realizations using different R(ϕ).Source: C. Vehicle-Mounted Antenna Modelling · p. 10, Figure 10 · p. 10
  • Roof-mounted V2X antennas generally have smaller Q, while bumper- and glass-mounted antennas tend toward larger Q or lower G0.The model is demonstrated and validated with V2X antennas; extension to other frequency bands and wireless functions is left for future work.Source: C. Vehicle-Mounted Antenna Modelling · p. 10

V. CONCLUSIONS

The paper establishes a large-scale full-vehicle measurement dataset and characterizes how installation affects antenna patterns across platforms, locations, and wireless functions. It also develops a physically interpretable statistical model whose generated patterns preserve dominant measured characteristics.

V. CONCLUSIONS

  • Measurements across ten commercial vehicle platforms, multiple mounting locations, and 4G, 5G, V2X, and GNSS show that integration reshapes patterns through blockage and structural asymmetry.Observed effects include directional imbalance, localized radiation nulls, and function-dependent coverage characteristics.Source: §V · p. 10
  • Statistical analysis using HPR, PAG, UHE, and horizontal fluctuation metrics reveals distinct radiation characteristics across wireless functions and installation configurations.Source: §V · p. 10
  • The proposed statistical model generates realistic patterns with a small set of physically interpretable parameters and preserves dominant characteristics of measured patterns.The paper presents the dataset and characterization as an empirical foundation for realistic system-level simulation and system design in intelligent transportation applications.Source: §V · p. 10
Loading 2609.36643v1…