Analysis of C-V2X Antenna Performance on Vehicular Panoramic Glass

,


Introduction
Cellular vehicle-to-everything (C-V2X) communication is crucial for autonomous driving and intelligent transportation systems.It leverages the evolution in the cellular systems from 4G to 5G cellular networks to enable reliable communication with low latency for vehicular networks.C-V2X has four applications which are vehicle-to-vehicle (V2V), vehicle-topedestrian (V2P), vehicle-to-infrastructure (V2I), and vehicleto-network (V2N) communications.By exchanging messages between vehicles, pedestrians, and infrastructure, C-V2X can support many use cases such as forward collision warning, vehicle platooning, and remote driving [1].
Antenna placement on vehicles signifcantly impacts the antenna performance, so most antennas are placed on the vehicle's roof to avoid signal blockage by the vehicle's metallic parts [2].Te continuous evolution of car aesthetic designs led to the introduction of many car roof types, such as sunroofs and panoramic roofs, to allow light and air inside the car [3].Lately, the automakers have introduced a new roof type, the full glass roof, to present a new experience and view for the passengers.However, introducing glass and plastic parts on the vehicle's roof afects the roof-mounted antenna performance [4].
Te goal of this paper is to analyze the performance of a C-V2X quarter-wavelength monopole antenna when mounted on a vehicle's glass roof through a series of measurements and simulations.Currently, there are no studies on the antenna performance when mounted on a vehicle's full glass roof in any frequency band.Few references discuss the C-V2X antenna performance when mounted on a metal behind a panoramic roof [4][5][6][7].Although the C-V2X antenna is mounted on the metallic portion of the vehicle's roof, the antenna performance is degraded to the front of the vehicle which is the direction of the panoramic roof area [4][5][6][7].
In this work, the performance degradation of the C-V2X roof-top antenna when mounted on vehicle's glass roof is investigated.Te measurements setup and the C-V2X monopole antenna measurements on the glass roof are presented in Section 2. Section 3 shows the C-V2X halfwavelength dipole antenna measurements.Section 4 discusses the C-V2X monopole antenna measurements when mounted on four diferent glass types.Section 5 presents the C-V2X monopole antenna simulation results.Te SAR study for the passengers inside the vehicle is presented in Section 6, and Section 7 concludes this paper.

C-V2X Monopole Antenna Performance on
Vehicle's Glass Roof A small green cable with an SMA connector is then soldered to the antenna to act as the antenna feed.Te antenna and the PCB were mounted on a small shark-fn chassis with a size (LxW) of 117 × 67 mm 2 , as shown in Figure 2. Te C-V2X antenna gain measurements are conducted at Oakland University's Outdoor Automotive Antenna Range in Michigan, USA.First, the C-V2X monopole antenna is measured when mounted on the 1-meter rounded metallic ground plane and then when mounted on the vehicle's glass roof, as shown in Figure 3. Te far-feld measurement data are analyzed and expressed in two parameters, i.e., radiation pattern (RP) and linear average gain (LAG).Te RP is measured at three elevation angles of (10, 0, and −6), corresponding to theta angles of 80 °, 90 °, and 96 °, respectively, at 5.9 GHz.Tese elevation angles are chosen to represent above-horizon (80 °), horizon (90 °), and below-horizon (96 °) angles.LAG is calculated by converting the measured gain values in dBi to linear values at each elevation angle between ⊖ � 80 °and ⊖ � 96 °, at each of the 360 azimuth points per elevation.Te linear values are then averaged and converted back to dBi to determine the LAG value for each frequency.For the antenna measurements, the angular resolution is 1 °and 2 °for theta and phi resulting in 16 theta points and 181 phi points per elevation point.
Figure 4 shows the performance comparison for the C-V2X monopole antenna when mounted on the metallic ground (GND) plane vs. mounted on the vehicle's glass roof.When mounted on metal, the antenna has an omnidirectional radiation pattern without nulls at all three elevation angles.In contrast, when the monopole was mounted on the glass roof, the radiation pattern had deep nulls on the sides at Ø � 90 °and 270 °.Te front and back directions of the radiation pattern do not show deep nulls like the side directions; however, the gain is signifcantly less than the metallic ground plane placement.Te antenna performance degradation when mounted on the vehicle's glass roof is evident in the LAG as shown in Figure 5, as the LAG for the monopole mounted on metal is about −1 dBi compared to −9 dBi when mounted on the glass roof.
To better understand the results in Figures 4 and 5, simulations of the C-V2X monopole antenna on the full glass roof were performed to investigate the reasons behind the performance degradation.Te Ansys high-frequency structure simulator (HFSS)' fnite element method (FEM) was used for the simulation with the glass roof CAD fles (Figure 6).Glass material was assigned a dielectric constant of 7 and a dielectric loss tangent of 0.009 [7].Te simulation results do not match the measurements shown in Figure 7 as the simulated RP shows void of deep nulls, which are missing in measured results, and simulated LAG shown in Figure 8 is more than eight dBi higher than the measurements, which indicates that there are materials other than the glass causing the antenna performance degradation on the full glass roof.

C-V2X Dipole Antenna Performance on Vehicle's Glass Roof
Although the C-V2X monopole antenna is the most used antenna in the automotive industry, a half-wavelength C-V2X dipole antenna is also added to the measurements campaign to study the efect of the vehicle glass roof on diferent antenna types.Te dipole antenna is suitable to be investigated as it is less dependent on the ground plane than the monopole antenna.Te C-V2X dipole antenna is measured when the dipole antenna is mounted on1-meter rounded metallic ground plane and then when mounted on vehicle's glass roof as shown inFigure 9. Te far-feld measurement data are analyzed and expressed in two parameters, i.e., radiation pattern (RP) and linear average gain (LAG) as in the monopole antenna case which was discussed earlier.
Figure 10 shows the performance comparison for the C-V2X dipole antenna radiation pattern when mounted on the metallic ground plane vs. when mounted on the vehicle's glass roof.When mounted on metal, the antenna has an omnidirectional radiation pattern without nulls at all three elevation angles.In contrast, when the dipole antenna was mounted on the glass roof, the radiation pattern had deep nulls.Te front and back directions of the radiation pattern do not show deep nulls like the side directions; however, the gain is less than the metallic ground plane placement.Te antenna performance degradation when mounted on the vehicle's glass roof is evident in the LAG shown in Figure 11, as the LAG for the dipole mounted on metal is about 1 dBi compared to −6.5 dBi when mounted on the glass roof.Te C-V2X dipole antenna followed the same trend as the C-V2X monopole antenna when mounted on the vehicle's glass roof which indicates that the antenna performance degradation when mounted on the vehicle's glass roof is not related to the antenna type.

C-V2X Monopole Antenna Measurements on Different Glass Types
As it is quite clear that the glass properties afect the antenna performance, diferent studies were performed and presented in this section to learn further the efect of roof material properties on antenna performance.Te C-V2X monopole antenna was measured on four diferent glass types which are as follows: (i) Clear glass sheet with the size of 508 × 508 × 3.2 mm 3 (ii) Low-emissivity (low-E) glass sheet with the size of 508 × 508 × 3.2 mm 3 (iii) Metal wire-reinforced glass sheet with the size of 585 × 585 × 6.35 mm 3 (iv) Vehicle's glass roof sample Figure 12 shows the measurement setup for the C-V2X monopole antenna when mounted on the four types of glass.Figures 13 and 14 show the radiation pattern and LAG for the C-V2X monopole when mounted on the four glass materials, respectively.When the antenna was mounted on clear glass, it had an omnidirectional radiation pattern, while the wire-reinforced glass and the vehicle's glass roof showed deep nulls in the radiation pattern, especially at ⊖ � 90 °and 96 °.Te antenna showed an omnidirectional pattern for the low-E glass case but with a gain less than the clear glass antenna mounting.Te antenna LAG shown in Figure 14 proves that when the antenna is mounted on clear glass, it performs well, like when mounted on metal.Te antenna mounted on wire-reinforced glass and glass roof exhibited the worst LAG performance due to the nulls in their radiation patterns.
Figure 15 shows the average gain over elevation angles for the C-V2X monopole antenna mounted on the four glass materials.For the antenna mounted on clear glass and low-E glass, the gain for elevation angles above the horizon (⊖ < 90 °) and below the horizon has comparable values, while the wire-reinforced glass and the glass roof showed better gain values in the above-horizon angles compared to the below-horizon angles.
Tese measurements conclude that the glass material is not the reason for the C-V2X monopole antenna  International Journal of Antennas and Propagation performance degradation since the clear glass shows a good performance.However, the antenna mounted on wirereinforced glass shows a trend similar to that of when the antenna is mounted on a glass roof, indicating that there may be some metallic coating or metallic particles in the construction of the vehicle's glass roof.

C-V2X Monopole Antenna Simulation on a Full Glass Roof
Te objective of this section was to get simulation results that matched the measurements for the C-V2X monopole on a glass roof.Te antenna measurements on the four glass materials showed the possibility of a metal layer in the glass roof sample, so the best assumption was that there is a low-E coating underneath the glass roof to protect the vehicle interior and passengers from the heat of the sun.Te glass roof CAD fles contained opaque areas at the edges of the glass roof while the middle of the glass roof sample was transparent, so we assumed that the low-E coating layer was in the opaque area at the edges while the transparent area was just pure glass.Te low-E coating is a thin transparent coating that is put on the glass to prevent heat and allow the passage of the visible light.One of the low-E coating simplest layer stacks is as follows: glass-SnO 2 (40 nm)-silver(10 nm)-SnO 2 (40 nm) [8,9].Te tin-oxide (SnO 2 ) layer is a high refractive layer which has a dielectric constant of 24 and a conductivity of 2 S/m [10,11].Figure 16 shows the layer stack used for the simulation.Te antenna is mounted on the glass layer, and the low-E coating is underneath the glass, with each layer having a thickness of 1 µm to avoid meshing errors during the simulation when lower thickness values are used.
Figure 17 compares the measured and simulated radiation pattern for the C-V2X monopole when mounted on the full glass roof sample.A good agreement is achieved between the simulation and measured radiation pattern, as the simulated radiation pattern shows deep nulls with slight diferences from the measurements.For ⊖ � 96 °, the simulated radiation pattern had more refections in the front direction compared to the measurements.Te LAG shown in Figure 18 also agreed well with values of −10 dBi and −9dBi, for the simulations and measurements, respectively.Figure 19 shows the simulated and measured average gain over elevation angles for the C-V2X monopole mounted on the glass roof sample.For the elevation angles above the horizon (⊖ < 90 °), a good agreement was achieved between measurements and simulation.However, for the below-horizon elevation angles, a diference of more than 1.5 dBi is observed From the simulation results, the main reason for the C-V2X antenna performance degradation on the glass roof sample is that the C-V2X electromagnetic wave propagates through diferent layers with diferent dielectric constants, which causes the wave to be refected and scattered.Te refections that appeared in the measured radiation pattern can prove this analysis.To reduce the wave refections, the low-E coating can be installed on the upper side of the glass  International Journal of Antennas and Propagation to be between the antenna and the glass layer, as shown in Figure 20.An HFSS simulation is performed for the C-V2X antenna when mounted on a glass roof with the low-E coating on the upper side of the glass.Figure 21 shows the C-V2X monopole simulated radiation pattern with the low-E coating on the top side of the glass roof.Te antenna had a good omnidirectionality in its radiation pattern at ⊖ = 80, 90, and 96 °without any nulls.Also, the LAG shown in Figure 22 was around −1.5 dBi, which is 8 dB better than when the low-E coating is on the bottom side.Tis fnding shows that moving the low-E coating to the top layer of the glass is crucial for the C-V2X antenna performance improvement on a glass roof as the antenna performance when the low-E coating is placed on the top of the glass layer is comparable to when the antenna was mounted on a metallic roof.

SAR Study of the C-V2X Antenna on Glass Roof
With the presence of the vehicle's glass roof, the passengers inside the vehicle will be exposed to the radiation from the C-V2X antenna mounted on the glass roof.To check the C-V2X antenna's radiation efect on the vehicle's passengers, a specifc absorption rate (SAR) analysis is conducted when the passengers are in the front and rear seats of the vehicle and the C-V2X monopole antenna is mounted on the top of the glass roof.
A simulation is conducted for the SAR using Ansys HFSS at the 5.9 GHz.Te SAR simulation setup is shown in Figure 23 as the C-V2X monopole antenna is mounted on the glass roof with the low-E coating on the top of the glass, as shown in Figure 20.Te front and rear passengers are included in the simulation.Te C-V2X antenna signal power is set to 23 dBm (0.2W) which is the maximum power that can be fed to a C-V2X antenna according to the SAE J3161 standard [12].Te human body dielectric constant is 45.8 and conductivity is 0.77 S/m [13].
Te HFSS SAR simulation shows that the maximum SAR value on the passengers is 0.007 W/kg which is negligible

Conclusions
Te performance degradation of the C-V2X monopole antenna when mounted on a vehicle's full glass roof section is analyzed through a series of measurements and simulations.Measurement of the C-V2X monopole antenna when mounted on vehicle's glass roof shows a performance degradation in the LAG of 8 dB compared to when the monopole antenna is mounted on 1-meter metallic ground plane.A half-wavelength C-V2X dipole antenna is then measured on the glass roof and on the 1-meter metallic ground plane.Te dipole antenna measurements on the glass roof showed similar performance degradation as the C-V2X monopole antenna which indicates that the performance degradation is not related to the antenna type.Te C-V2X monopole antenna is then measured on four different glass types in which the monopole antenna showed degraded performance when mounted on the wirereinforced glass indicating the possibility of the existence of metallic coating or metallic particles inside the glass roof section structure.Te simulation showed that the performance degradation is caused by the low-E coating layer underneath the glass in the opaque area at the edges of the glass roof, which caused refections for the C-V2X electromagnetic waves.Te simulation shows that if the low-E coating is moved from underneath the glass to the top, the antenna performance will improve signifcantly.A SAR study was conducted and showed that the passengers are exposed to negligible radiation from the C-V2X monopole antenna mounted on the glass roof.

Figure 1
Figure1shows a vehicle's glass roof section provided by an automotive original equipment manufacturer (OEM) with its computer-aided design (CAD) fles.No other information is provided by the OEM or the glass supplier about the construction of the glass roof or the layers stacked inside the glass.Te glass roof section consisted of a metal frame underneath the glass and the rear mirror on which the antenna was mounted, as shown in Figure1.Te glass size (L × W × H) is 982 × 620 × 3.85 mm 3 with the antenna mounting hole located 125 mm from the back end of the glass.Due to the curvature of the glass roof, the antenna is tilted by an angle of 8.3 °in the yz plane as the front of the glass roof is higher than the back of the glass roof.Also, the sides of the glass roof are curved in the xz plane as shown in Figure1.Te quarter-wavelength C-V2X monopole antenna is used in this study as the monopole is the most used C-V2X antenna in the automotive industry.Te monopole antenna has a height of 13 mm and a width of 2 mm as shown in Figure2.Te antenna is soldered to a printed circuit board (PCB) with an FR4 material which has a dielectric constant of 4.4.A small green cable with an SMA connector is then soldered to the antenna to act as the antenna feed.Te antenna and the PCB were mounted on a small shark-fn chassis with a size (LxW) of 117 × 67 mm 2 , as shown in Figure2.Te C-V2X antenna gain measurements are conducted at Oakland University's Outdoor Automotive Antenna Range in Michigan, USA.First, the C-V2X monopole antenna is measured when mounted on the 1-meter rounded metallic ground plane and then when mounted on the vehicle's glass roof, as shown in Figure3.Te far-feld measurement data are analyzed and expressed in two parameters, i.e., radiation pattern (RP) and linear average gain (LAG).Te RP is measured at three elevation angles of (10, 0, and −6), corresponding to theta angles of 80 °, 90 °, and 96 °, respectively, at 5.9 GHz.Tese elevation angles are chosen to represent above-horizon (80 °), horizon (90 °), and below-horizon (96 °) angles.LAG is calculated by converting the measured gain values in dBi to linear values at each elevation angle between ⊖ � 80 °and ⊖ � 96 °, at each of the 360 azimuth points per elevation.Te linear values are then averaged and converted back to dBi to determine the LAG value for each frequency.For the antenna measurements, the angular resolution is 1 °and 2 °for theta and phi resulting in 16 theta points and

Figure 16
Figure 16: C-V2X monopole simulation on a full glass roof (a) simulation setup and (b) glass roof simulation stack up.

Figure 18 Figure 19
Figure 18: C-V2X monopole antenna measured and simulated LAG when mounted on a full glass roof.

Figure 20 0
Figure 20: C-V2X monopole simulation on a glass roof with the low-E coating on the top (a) simulation setup and (b) glass roof simulation stack up.