Gain-enhanced petal-shaped MIMO antenna system with FSS loading for sub-6 GHz V2X communications

Abstract

In this research work, a compact dual-port petal-shaped MIMO antenna printed on FR4-substrate with volume of 24 × 24 × 0.787 mm³ is presented for vehicle-to-everything (5.85–5.95 GHz) wireless communication. The gain-enhancement is also achieved by loading frequency-selective-surface array (FSS_V2X) of size 204 mm×204 mm printed on one surface of FR4 substrate with 1.60 mm thickness which records measured peak-realized-gain of 7.58 dBi within the operating-bandwidth of 4.12–6.92 GHz. High measured isolation of more than 25.0 dB is also achieved by placing multiple discontinuous rectangular strips between ground by an angle of 45°. The permissible diversity parameter also records excellent performance with ECC_V2X<0.50, DG_V2X>9.95 dB, TARC_V2X<0.0 dB and CCL_V2X<0.40 b/s/Hz. The above features of FSS-loaded MIMO antenna is a good candidate for vehicle-to-vehicle automotive (V2V), vehicle-to-pedestrian (V2P) and vehicle-to-infrastructure (V2I) applications.

Introduction

The rapid evolution from 5G toward emerging 6G technologies, combined with the proliferation of the Internet of Things, and connected vehicles, has intensified the need for seamless integration of vehicular 5G antennas into modern automotive platforms, particularly for Vehicle-to-Everything (V2X) communication. Enabled by the Internet of Vehicles (IoV), such connected systems facilitate real-time data exchange via mobile networks and have become one of the most dynamic areas of research in wireless communication. Current V2X communication standards, based on IEEE 802.11p, operate in the 5.850–5.92 GHz frequency range and support diverse applications such as vehicle-to-networks (V2N), vehicle-to-infrastructure (V2I), vehicle-to-vehicle (V2V), and vehicle-to-pedestrian (V2P) communication^1,2. These capabilities allow vehicles to share critical information on safety alerts, weather updates, traffic conditions, and hazardous driving patterns^3,4. The integration of cellular V2X (C-V2X) with 5G technologies further enhances these systems by offering low latency, improved road safety, efficient traffic management, and better coordination between vehicles and pedestrians. As intelligent transportation systems (ITS) continue to expand, the demand for antennas that can provide high throughput, reliable connectivity, and minimal latency is becoming increasingly significant. In this context, multiple-input-multiple-output (MIMO) antenna architectures have emerged as a key enabler for achieving robust, high-performance V2X communication, positioning themselves at the forefront of next-generation vehicular communication research⁵. To realize such connectivity lies the core challenge in designing antennas with a low profile, wideband response, good radiation behaviour, while maintaining the spatial constraints of automotive systems. Multiple-input multiple-output (MIMO) antenna configurations signify their contributions in this regard by providing improved data throughput and reliable link performance. However, the critical requirements for the design of MIMO antenna configurations for automotive applications are to ensure that sufficient inter-element isolation and improved bandwidth^6–9 without altering their compact footprint.

Several strategies have been explored for achieving bandwidth enhancement and increased isolation in compact MIMO antenna configurations. Microstrip Patch Antennas (MPA) comply with planar configuration, fabrication simplicity, and better printed circuit boards (PCBs) integration capability. But at the same time, these MPAs suffer from narrow band operation and high inter-element spacing⁶. Recently, numerous isolation and decoupling methods have been utilized that include neutralization line^10–12, defective ground structure (DGS)^13–15, decoupling network^16–19, use of parasitic elements^20,21, etc., near field resonant structures^22–24 with different levels of success. On the other hand, Dielectric resonator antennas (DRA) show promising gain and efficiency, but at the cost of high fabrication cost and design complexity. Thus, achieving a compact antenna with wideband operation exhibiting high isolation within the spatial constraints of automotive platforms remains a significant challenge.

To address these challenges, Frequency Selective Surface (FSS) have gained attention in enhancing the radiation performance of MIMO communication systems in recent years^25,26. These FSS acts are periodic structures that can manipulate the electromagnetic waves to modify their behaviour to enhance antenna radiation characteristics, such as enhancing isolation, bandwidth, radiation gain, and polarization. The work proposed in²⁷ outlines the integration of a meta-surface along with a patch antenna configuration to enhance radiation characteristics, but exhibits high design intricacies and a larger footprint. The work discussed in²⁸ highlights the 68-unit SCS-shaped FSS array integrated with a chair-shaped MIMO antenna design, along with utilizing parasitic elements to achieve a high gain of 7.96 dB with good isolation levels. Further, the efficacy of the FSS layer having Jerusalem Cross (JC) shaped structure with 8-port MIMO antenna in²⁹ achieves excellent wideband performance with good isolation values. Another work presented in³⁰ signifies the utility of FSS in boosting the antenna gain with a compact form factor and high radiation efficiency. Thus, these works validate the contribution of FSS to achieve decoupling performance and suppress surface waves in addition to enhancing the antenna’s radiation performance while maintaining the antenna profile.

In parallel, Characteristic Mode Theory (CMT) has emerged as a significant analytical tool to optimize the performance of MIMO antennas and realize the enhanced bandwidth and isolation characteristics with better radiation behaviour. The CMT framework, unlike adopting a simulation-based technique, offers researchers to choice of beneficial resonant modes by providing a conceptual, intuitive interpretation of the associated modal behaviour^31–33. Also, CMT enables the decoupling of different characteristic modes and selectively combines the modes lying near the resonant frequency to achieve wideband behaviour. The work proposed in³¹ shows the significance of implementing CMT in achieving broadband operation and a good isolation level (> 15 dB) by analysing the modal behaviour. Further, the work discussed in³² details the utilization of CMT in identifying radiant and non-radiant modes and placement of the feedline. Further, the work demonstrated in³³ highlights the contribution of CMT in mapping the mode-to-radiation behaviour in designing a tuneable and efficient MIMO system. Thus, it is concluded that CMT definitely provides physical insights about the antenna behaviour along with the modal excitation process to optimize the performance of MIMO systems across a wide operational frequency range. A four-port MIMO antenna fabricated on Rogers substrate generates bandwidth between 3.0 GHz and 6.0 GHz and the integration of FSS achieves maximum measured peak-gain of 4.80 dBi³⁴ and also, modified circular patch-antenna generating ultra-wideband (3.30 GHz-10.80 GHz) is integrated with FSS recording maximum peak-gain of 8.0 dB³⁵. An Automotive communication MIMO-antenna is designed for multiple lower-band wireless application in Ultra-wideband maintaining isolation of more than 20.0 dB³⁶ and also MIMO-antenna with multiple merged elliptical structures³⁷ is also useful for ultra-wideband applications with orthogonal arrangement measuring isolation of more than 20.0 dB is verified for automotive applications. Four-port MIMO-antenna exhibiting peak gain of 11.30 dBic generates bandwidth of 5.40 GHz-6.30 GHz is designed for V2X communication^38–41.

Although the proposed antenna employs established concepts such as petal-shaped radiating elements, defected ground structure (DGS)-based isolation, and an FSS reflector, the fundamental novelty of this work lies in their systematic integration and co-optimization for multi-band V2X MIMO applications. Unlike recent V2X MIMO antenna designs that typically focus on either radiator shaping or isolation enhancement alone, the proposed design achieves simultaneous multi-band operation, high isolation, and gain enhancement through a unified design framework. In particular, the mutual interaction between the petal-shaped radiator and the connected DGS is exploited to improve impedance matching and inter-element isolation without additional decoupling networks, while the FSS reflector is optimized to enhance gain without disturbing the MIMO characteristics.

This paper proposes a compact two-port petal-shaped MIMO antenna exhibiting a wide impedance bandwidth (1.9 GHz) for 5 G-enabled V2X communication applications. The proposed MIMO configuration incorporates an innovative isolation mechanism utilizing a slotted arrow-shaped self-decoupling network to enhance performance. The designed MIMO antenna is etched using a cost-effective FR4 substrate and is easy to mount roof (top) of vehicles. The designed antenna is integrated with FSS enhances its gain and radiation efficiency, and also the V2X antenna is experimentally characterized to validate its effectiveness over the desired operating spectrum. The objective of this study are as follows:

1. A low-cost compact, dual-element petal-shaped MIMO antenna system exhibiting wideband operation (5.1–7.3 GHz) that includes 5G sub-6-GHz band (4.40–5.00 GHz) with sufficient inter-element isolation ensuring good radiation behaviour, is proposed for 5G-enabled V2X communication band (5.8–5.9 GHz) applications.

2. The bandwidth of the proposed MIMO antenna is enhanced by implementing a modified petal-shaped monopole structure. Further, the isolation is enhanced by replacing the ground plane with a decoupling network. This eliminates the need for an external decoupler network to achieve high isolation and efficiency.

3. The designed MIMO antenna has a small footprint of size 24 × 24 × 0.8 mm³, justifying its suitability for vehicular applications. Further, a comprehensive experimental analysis is done to evaluate MIMO performance metrics, including Envelope Correlation Coefficient (ECC), Mutual Coupling Loss (MEG), Cross-correlation Loss (CCL), Total Active Reflection Coefficient (TARC), and spatial multiplexing efficiencies.

The work is organized in different sections as follows:

Section “Single-element antenna design, and its equivalent circuit model analysis” discusses the design methodology of a single antenna element, along with its evolution steps and performance. The Section “Frequency-selective-surface (FSS) for V2X communication” discusses the single-element FSS which will be loaded with the antenna for gain-enhancement and is considered to be one of the technique for achieving high gain in planar technology. Next, its extension to the dual-port MIMO antenna system design is implemented in Section “Two-port MIMO antenna design”, followed by its simulated performance evaluation in Section “Diversity Performance of the two-port MIMO system (ECCV2X, DGV2X, TARCV2X, CCLV2X, MEGV2X)”, and finally, followed by a conclusion.

Single-element antenna design, and its equivalent circuit model analysis

Initially, a petal-shaped monopole antenna is designed to achieve resonance for V2X communication applications in the sub-6 GHz spectrum. The antenna is fabricated by etching 0.035 mm-thick copper on a 0.787 mm-thick FR4 substrate (εr = 4.4, tan δ = 0.025). It is excited using a conventional microstrip feedline, facilitating easy integration with planar RF circuitry. The overall dimensions of the antenna are 8.00 × 10.70 × 0.787 mm3. Figure 1 illustrates the geometry of the designed antenna, with the optimized parameters provided in Table 1. Additionally, Fig. 1a shows the front view of the geometry, and, Fig. 1b shows defected ground structure of dimensions W × Gl mm2, which aids in achieving a narrow impedance-matched bandwidth suitable for V2X applications.

Fig. 1.

One-port V2X antenna. (a) Top and, (b) bottom view.

Table 1.

Optimized antenna parameters.

Dimension	mm	mm	Dimension	mm
L	14.0	24	Fl	2.5
W	10.0	12	Fw	3.12
Hx	8.00	8	Gl	2.5
Hy	10.70	13

Figure 2 illustrate the ECM analysis of one-port antenna. Figure 2a depicts the real-impedance graph where the net impedance at 5.90 GHz corresponds to (46.07−j0.47) Ω. This value of impedance shows that at resonance of 5.90 GHz, the real value is near to ideal value of 50 Ω, whereas the imaginary value corresponding to 0.47 is near to ideal imaginary value of 0 Ω with a negative sign indicating the mild capacitive nature. Figure 2b shows the modelling of the patch antenna with equivalent circuit model comprising of feed-line impedance and impedance of the patch. The feedline is modelled as series connected R (14.5 Ω) − L (0.795 nH) circuit with C (1pF) connected in parallel. The patch circuit is modelled as parallel RLC circuit^38–41, where the values are calculated from Eqs. (1)–(3) given below.

1

2

3

Fig. 2.

Equivalent-circuit-model (ECM) analysis. (a) Real-imaginary graph. (b) Screenshot of ECM in EM-circuit simulator ADS (c) S-parameters for the optimized petal-shaped single antenna element. (d) S₁₁ comparison.

The value of R = 46.07Ω, and from the imaginary value, the inductance is calculated, which is substituted in Eq. (3), and the value of capacitance is calculated. The values obtained are used in realizing the parallel resonance circuit, which is sketched in the EM-circuit simulator ADS as shown in Fig. 2b. The port is terminated by 50 Ω, and a sweep of 4.00–8.00 GHz is applied with a step size of 0.001 GHz. The simulated S11 extracted from the EM-circulator simulator is shown in Fig. 2c with a screenshot of the resonance value corresponding to 5.90 GHz. The comparison of S11 extracted from EM-simulator and from circuit-simulator is compared in Fig. 2d with simulator bandwidth corresponding to 5.056–7.452 GHz, with resonance frequency = 5.828 GHz (S₁₁ = − 37.57 dB), and circuit-simulator bandwidth of 5.8672–5.967 GHz with resonance = 5.9169 GHz (S₁₁ = − 27.7639 dB).

In practical vehicular environments, antenna performance can be affected by factors such as vehicle body loading, installation location, manufacturing tolerances, and nearby electronic components, which may cause frequency detuning. A wide bandwidth ensures stable operation within the intended V2X band under such non-ideal conditions.

Moreover, next-generation V2X platforms are increasingly expected to support multi-standard and multi-service operation, including 5G sub-6 GHz, WLAN, and future vehicular communication extensions. The wide bandwidth allows a single antenna system to accommodate these services without redesign, thereby reducing system complexity and cost.

Evolutionary approach to single-element antenna design

The proposed antenna design has evolved based on various iteration processes, with every step targeting better impedance matching with enhanced bandwidth for the desired V2X band, as shown in Fig. 3, with evolution and the extracted S-parameter result.

Fig. 3.

Evolution of one-port antenna.

In step 1, the basic elliptical monopole structure was initially developed as depicted in Fig. 3a, showing resonance at 8.9 GHz with a return loss of − 21 dB and a narrowband response Fig. 3b. Subsequently, to shift the resonance towards the desired frequency of 5.9 GHz, structural modifications were made by tailoring the top right and bottom left sections of the initial elliptical geometry, yielding a dual-resonance behaviour at 7.8 GHz and 9.5 GHz, respectively. Further, these tailored sections were smoothly reshaped into curved profiles forming a petal-shaped structure. This alteration resulted in increased effective electrical length for the current, facilitating resonance at the 5.9 GHz with enhanced bandwidth covering the V2X band (GHz) and appreciable return loss (37.22 dB) in alignment with 5G applications requirements.

Parametric analysis of single element antenna design

The parametric analysis is performed to optimize the dimensions of the designed antenna element. The feedline width and the height of the ground are tuned to achieve the resonance at the desired frequency 5.9 GHz.

The ‘feed width’ is varied from 0.858 to 1.058 (in steps of 2 mm) and the ground height ‘gvary’ is varied from 2 to 3 mm (in steps of 0.5 mm) respectively, as shown in Fig. 4a,b. The corresponding S-parameters are plotted as shown in Fig. 2. Figure 4a illustrates that varying the height of the ground results in the variation of impedance matching. Thus, the optimum ground height is found to be 2.5 mm. Further from Fig. 4b, it is depicted that as the feed width is increased, the resonance is shifted towards the left. Also, the impedance matching improves with an increase in the dimensions of the feed width. The optimized dimensions of the feed width are chosen to be 1.058 mm.

Fig. 4.

Parametric Variation of S₁₁ plot w.r.t. (a) ground height, (b) feed width.

Radiation performance of single element antenna design

The simulated gain performance of the single port antenna is analysed to investigate the radiation performance, as shown in Fig. 5. The antenna radiates well with consistent gain performance across the operational frequency band, ranging from 5.2 to 7.10 GHz, maintaining wideband performance. The gain is observed to be 1.23 dB at a centre frequency of 5.9 GHz, with a peak value of 1.35 dB in the upper frequency range, justifying its suitability for V2X communication applications while maintaining a good trade-off between gain and bandwidth as recorded from Fig. 5a. Also, the radiation efficiency is recorded in Fig. 5b with a stable efficiency of 71% recorded around 5.9 GHz.

Fig. 5.

Gain versus frequency for the proposed single element.

Further, the surface current distribution to illustrate the radiation behaviour of the single-port antenna at 5.9 GHz is shown in Fig. 6. The current distribution has a high concentration around the microstrip feed region, exciting uniformly and symmetrically the broader region of the petal-shaped structure. This validates the maximum transfer of energy from the Feedline to the radiating patch structure. Also, the strong concentration of current near the bottom region indicates good coupling between the feed and the radiating element.

Fig. 6.

Surface current distribution around the proposed single-element antenna.

Frequency-selective-surface (FSS) for V2X communication

In this section Single port FSS and FSS array for V2X communication is discussed in detail.

Single-port frequency-selective-surface for V2X communication

The frequency-selective surface (FSS) is broadly defined as the periodic structures fabricated on a dielectric with conductive patches printed on one of the surfaces. The FSS uses the principle of resonance along with the periodicity, where the unit cell of the FSS is designed to work for the required applications. The two types of FSS are classified as band-pass and band-stop, where the former allows the range of frequency to pass while blocking others, and the latter blocks the range of frequency while allowing others.

The FSS is broadly found its applications in antenna design to increase gain, radomes (protective 3D cover for antennas), stealth technology to reduce radar-cross-section (RCS), and filters/absorbers.

The transmission and reflection coefficients can be correlated with admittance, with Ao corresponding to free-space admittance, which is the reciprocal of intrinsic impedance (ηo = 377Ω), and As is the equivalent surface admittance of the FSS.

The reflection coefficient (RC) is defined as

4

And coefficient (TC) is given by

5

The designed FSS, working as a band-pass or band-stop at resonance, is determined by the condition As A0 (band-pass) and As Ao (band-stop).

However, the resonance frequency, fr for FSS, is given by

6

where C is the speed of light (3 × 10⁸ m/s), F₁ is the length and width of the unit-cell element, and is effective dielectric constant.

Figure 7 illustrates the detailing of the unit-cell FSS with evolution and Table 2 shows the optimal dimensions which is designed to resonate at 5.90 GHZ for V2X wireless communication. Figure 7a shows the Stage1 of the FSS printed on FR4 dielectric substrate with conducting ring structure. The S-parameter plot shows that the FSS is suitable for resonance frequency at 1.42 GHz. However, the objective of the proposed work is targeted to design the FSS for V2X communication (5.90 GHz). Hence, the next stage, Stage2 as shown in Fig. 7b is developed where the rectangular ring is embedded with three horizontal-vertical strips and generates resonance at 4.62 GHz which is near to 5.90 GHz. The third stage, Stage3, records the addition of nine-circles of radius 1.50 mm which helps in shifting resonance to 5.31 GHz. The final version of the FSS, with addition of circle of radius 2.8125 mm achieves the objective of the FSS being usable at V2X communication which is desired. Figure 7c records the S-parameters (reflection and transmission coefficients) which generates band stop filter bandwidth of 2.387 GHz–8.562 GHz with resonance centered at 5.313 GHz (S12/S21 = − 49.85 dB). However, the required resonance must be at 5.90 GHz and hence, the shifting at 5.90 GHz is desired. Figure 7d shows the front-view of the FSS unit-cell design where the metal-patch is printed on FR4 substrate of thickness Fh mm with permittivity of 4.40. The printed-metal consists of a rectangular ring of dimensions F1 × F2 mm2 and a width of 2.10 mm. At the centre of the rectangular ring, a circular patch of radius Ra is printed, which is surrounded by eight smaller circles of radius Rb mm. The rectangular-ring and all the eight-circles are interconnected by rectangular strips of size L₁, L₂, L₃, w₁, W₂, W₃ and width of W mm. Figure 7e shows the 3D-view of the unit FSS-cell with smaller eight circles (C₁–C₈) placed near the centrally placed circle, C_o. Figure 7f shows the simulation-model of FSS placed within the closed boundary with two opposite faces touching the FSS edges assigned as E_t = 0 and the remaining pair assigned as H_t = 0. The top and bottom surface is excited by either port P1 or port P2. Also, Fig. 7g shows the S-parameter result of the excited unit-cell FSS. The FSS offers an operating bandwidth of 2.0–9.0 GHz. The S₁₁/S₂₂ grazes the 0.0 dB line, which indicates maximum reflection at both the ports (P₁ and P₂). Similarly, S₁₂/S₂₁ are below 20.0 dB for a bandwidth of 4.0–7.0 GHz, indicating very minimal signal is transmitted between P₁ to P₂ or P₂ to P₁. Hence, the maximum signal incident on FSS is almost reflected with minimum transmission and thus acts as a band-stop filter for the bandwidth between 4.0 and 7.0 GHz.

Fig. 7.

Unit-cell FSS. (a) Stage1. (b) Stage2. (c) Stage3. (d) Optimal-dimension. (e) 3D-view of unit cell. (f) The excitation model. (g) S11 results (reflection and transmission coefficients). (h) Oblique incident of plane wave at different angle of incidence.

Table 2.

Optimal dimensions of the FSS unit-cell.

OD	In mm	OD	In mm
Fa	34.0	Fb	34.0
F1	32.0	F2	32.0
W	1.90	W1	9.50
W2	9.50	W3 = L2	4.125
L1	1.70	Ra	2.8125
L3	9.50	Rb	1.50
Fh	1.60	d	25.0

OD, optimal dimension; mm, millimetre.

Figure 7h further shows the performance of the unit cell when the plane wave is incident at different angles of 0°, 15°, 30°, 45° and 60°. The values of angle corresponding to θ with ϕ = 0°, the S-parameter values are noted in Table 3 which concludes that the plane wave impinging at any angle do not deteriorate the performance of the unit cell making its array version making suitable to be integrated with the proposed V2X MIMO antenna.

Table 3.

Scanning angles.

Theta (θ)	Phi (ϕ)	V2X Resonance Frequency (GHz)	Smax (dB)	Comments
0°	0°	5.90	− 31.90	Normal incidence
15°	0°	5.90	− 32.08	Slight tilt
30°	0°	5.90	− 32.24	Moderate tilt
45°	0°	5.90	− 32.11	Severe tilt
60°	0°	5.90	− 32.01	Extreme tilt

FSS-array for V2X communication

The periodic structures are formed when an infinite array of identical unit cells is arranged in two dimensions, forming m × n (m = 2,3,…k; n = 2,3,…k). The arrangement of such an array with a 5 × 5 array corresponds to the dimension of Faa × Fbb = 204 mm × 204 mm shown in Fig. 8a. Also, the 3D view of the FSS structure is shown in Fig. 8b with a height of the FR4 substrate with height F_hh = 1.60 mm. The FSS-array is excited by a plane wave incident on the surface of the passive FSS-array. The plane-wave excitation includes the incoming plane wave (E_i), which divides into two parts with partial transmission in the forward direction, denoted by E_t, and the partially reflected, denoted by E_r. Under ideal conditions, the reflected wave (E_r) may be equal to the amplitude of the incident plane wave (E_i), and the transmitted wave (E_t) will be zero.

Fig. 8.

FSS-array. (a) Front-view of 5 × 5 FSS-array. (b) 3D view of the FSS-array prototype. (c) 2D-pattern. (d) 3D-pattern.

The FSS-array reflection coefficient (FSRC) is given by

7

And the FSS-array transmission coefficient (FATC) is given by

8

Figure 8c and d show the 2D plot of the FSS where a plane wave is incident, and it can be seen that the designed FSS-array shows the capability of a band stop filter, and the signals are reflected. These characteristics of the FSS in loaded with the MIMO antenna to enhance the gain, which is discussed in the upcoming sections.

Two-port MIMO antenna design

The initially designed 1-port petal-shaped antenna is expanded to achieve a dual-element MIMO antenna using the same FR4 substrate. To attain a MIMO configuration, two-petal-shaped monopole antenna elements are diagonally positioned on the top of the substrate, placed closely as observed in Fig. 9a. Further, as before, these diagonal elements are excited through a microstrip feed arrangement optimised to attain resonance in the desired V2X frequency band. The partial ground plane shown in Fig. 9b is incorporated into the bottom of the substrate to enhance its efficiency with improved impedance matching. The geometry of the chosen petal-shaped radiators contributes to enhanced effective aperture, leading to a broad impedance bandwidth. The resulting S-parameters are depicted in the graph shown in Fig. 9c. It is observed that the antenna exhibits restricted isolation of − 13 dB owing to strong near-field coupling between adjacent antenna elements. Figure 9d and e show the interaction of the inter-spaced element when one of the ports is excited and the other port is terminated by 50 Ω at 5.90 GHz.

Fig. 9.

Two-port MIMO. (a) Radiating view. (b) Ground view. (c) S-parameter results; Surface current distribution around the proposed two-port MIMO system when excited with. (d) Port 1 and, (e) Port 2.

Figure 9d shows the excitation of Port 1 at 5.90 GHz with maximum surface-current–density (SCD) accumulating within the transmission line. However, the radiating patch connected to Port 1 observes uniform distribution, and at the centre of the patch, the signals are radiated. The identical distribution of SCD is also observed in Fig. 9e with Port 2 excited and Port 1 terminated. In both cases, the neighbouring radiating elements’ interference can be observed, and hence, this interference needs to be reduced. This leads to the conclusion that some isolation mechanism is necessary for achieving the desired isolation performance as outlined in the next section.

Common-connected ground and isolation mechanism

The separated ground, which is shown in Fig. 9, is the case of non-connected ground. In MIMO-configuration, the different identical MIMO-elements with separated ground (not-connected) the directly split and provide direct enhancement of isolation, which is due to no-current coupling through the ground-plane as it is discontinuous. In a real system, the split of ground as shown in Fig. 9 is not practical because the signals should have a common-reference plane so that all the signal levels within the system can be analysed properly based on the zero-reference level. This misuse of not-connected ground is eliminated in the two-port MIMO antenna and is shown in Fig. 10. As discussed in the previous section, an effective isolation mechanism is essential to mitigate the coupled currents between closely spaced adjacent antenna elements in the proposed MIMO configuration for V2X communication. Therefore, a defective ground structure (DGS) is strategically embedded within its ground plane to address this issue, as shown in Fig. 10a.

Fig. 10.

Isolation enhancement. (a) Decoupling Structure having Diagonally oriented arrow-shaped DGS. (b) Isolation levels between the adjacent antenna elements.

This is governed by etching an array of diagonally oriented square-shaped slots to introduce a band-stop response that disrupts the propagation of surface currents between the adjacent ports, leading to enhanced isolation over the operational V2X band (5.85–5.925 GHz). Additionally, an arrow-shaped structure is integrated to extend the current path for incorporating additional discontinuities to suppress mutual coupling. Figure 10b shows the transmission coefficients between the two ports, i.e., S12 and S21, to show the isolation level between the two ports with and without a decoupling structure. From Fig. 10b, a significant improvement in isolation is observed, with a maximum value of − 20.5 dB through the combined implementation of these two structures, validating its suitability for 5G and V2X communication automotive systems. This improvement is realized without compromising its compact footprint and its radiation behaviour. Moreover, this decoupling network ensures a low signal degradation and minimal interference, which is critically desired for low diversity gain and better reliability for dynamic vehicular environments.

Surface current distribution, peak-gain, and radiation-efficiency analysis

Figure 11a,b depicts the current distribution on the surface of the designed tow-port petal-shaped MIMO antenna at 5.9 GHz. These two subplots illustrate the surface current distribution on the two MIMO ports at two-time instances, each corresponding to an orthogonal polarization state. In both plots, the results show strong current excitation by the feed near the central portion and along the edges of the petal-shaped patch elements when excited independently, confirming strong radiation behaviour by each MIMO element. Minimal coupling due to negligible current overlap between the two orthogonally placed elements signifies excellent isolation. Also, it is visible that when one port is excited (Fig. 11a), the second port (Fig. 11b) remains largely inactive and vice versa, demonstrating minimized mutual coupling. This isolation is obtained through a structural design feature of incorporating a decoupling network. The low surface current concentration on the non-excited element is desirable to preserve signal integrity, which helps to minimize the correlation between radiating elements and enhance MIMO performance.

Fig. 11.

Surface current distribution around the proposed two-port MIMO system with decoupled network when excited with (a) Port 1 and (b) Port 2 (c) 3-D model of the car-model with antenna placed on roof (d) Simulation of the car-antenna model.

Figure 11c also includes the 3-D car model analysis where MIMO-antenna placed on the roof-top where the body of the car-model is assigned the conducting material (perfect conductor). Figure 11d shows the simulated MIMO antenna placed on the roof-top of the car-model where the antenna maintains the omni-directional and the di-pole pattern in the principal planes catering the vehicle-to-anything communication. Figure 11e also record the S-parameters in the simulation model of car where MIMO antenna placed over the roof. The resonance frequency of 5.824 GHz is recorded for both the port excitation while maintaining the isolation of more than 18.0 dB in the operating bandwidth (5.064–6.998 GHz).

Additionally, the current minima along the diagonal spacing between the two antenna elements demonstrate the antenna’s capacity to minimize the interference and ensure better diversity gain and low ECC.

Figure 12 illustrates the design of a two-port MIMO antenna and the result analysis. The radiating patch with partial ground generated a narrow-band with a bandwidth of 5.05–7.452 GHz, and the resonance frequency centered at 5.828 GHz with S11 = − 37.57 dB noted from Fig. 2d. However, the single-port configuration, when deployed in a real-time environment, suffers from fading where the transmitted signals reach the receiver at different intervals of time due to reflections from multiple objects. To overcome the above-mentioned effects, the increase in the number of radiating elements from one to two increases the probability of receiving the signals more efficiently, and the technology is termed multiple-input-multiple-output (MIMO) with a spatial-diversity scheme. Figure 12a shows the two-port configuration MIMO-antenna where two-radiating elements RP1 and RP2 are printed on the top-surface and are placed orthogonally on FR4 substrate with dimensions of Wm × Lm = 27.60 mm × 27.60 mm. Also, the interspacing distance of I_S = 11.40 mm is maintained between the two radiation structures for better isolation.

Fig. 12.

The two-port MIMO antenna configuration (a) Front-view orientation details (b) Ground view dimension details (c), (d) Photograph of the prototype (e) Measured VNA screenshot (f), (g) Simulated-Measured S-parameters.

Figure 12b shows the ground where the isolation technique, achieving high isolation, is discussed in Fig. 10. The ground consists of two rectangular strips which are placed orthogonally to each other with total length of each strip corresponding to (Lm1 + Lm2) = (16.10 + 11.50) = 27.60 mm with thickness of t1,t2 corresponding to values of 1.15 mm, 1.495 mm respectively. A rectangular strip is placed diagonally with respect to the ground by an angle of 45°, which forms the isolation de-coupling structure with lengths L₁, L₂, and L₃ corresponding to 16.15 mm, 1.15 mm, and 2.9325 mm, respectively, with a thickness of W₁ = 0.92 mm. The rectangular strip is not continuous and includes nine broken strips with a gap between them of g₁ = 1.15 mm running through the diagonal as shown in Fig. 12b. The fabricated photograph with front-ground views is shown in Fig. 12c and d, where the photo-lithographic method of fabrication is used to achieve high precision for the accuracy of results. The S₁₁ is measured using a vector-network-analyser (VNA) with a screenshot shown in Fig. 12e, which records the sharp resonance at 5.90 GHz and is applicable for V2X communication. The simulated and measured S-parameters are shown in Fig. 12f and g. The simulated S₁₁/S₂₂ shows the overlapping on one with − 10.0 dB bandwidth of 5.152 GHz-7.08 GHz with resonance centred at 5.924 GHz for radiating-patch RP1, and for radiating-patch RP2, the resonance value corresponds to 5.936 GHz with S₁₁ = − 54.62 dB. The arrangement of MIMO-antenna elements achieving spatial-diversity sharing a commonly-connected ground maintains isolation of more than − 17.64 dB in the operating bandwidth. Figure 12g illustrates the measurement of S₁₁/S₂₂ and S₁₂/S₂₁ parameters of the prototype shown in Fig. 12c and d. The deviation in both coefficients and transmission coefficients is observed due to practical imperfections such as precise fabrication of the prototype with exact dimensions, variation in material property (permittivity = 4.40 and loss tangent), parasitic effects introduced by SMA connectors, and calibration errors. However, the measured results shown in Fig. 12g are in close agreement with the simulated results with measured S₁₁ (RP1) − 10.0 dB bandwidth of 4.80–6.82 GHz and the maximum resonance centred at 5.94 GHz (S₁₁ = − 28.38 dB). Also, the radiating-patch, RP2, generates an operating bandwidth of 4.12–6.92 GHz with resonance at 5.90 GHz (S₂₂ = − 48.35 dB). The frequency shift between RP1 and RP2 in a MIMO antenna, despite identical geometries, commonly occurs due to practical measurement and environmental factors rather than design flaws. Mutual coupling between elements alters current distribution differently for each excited port. Variations in SMA soldering, connector length, cable bending, and fabrication tolerances of FR4 substrate also introduce slight impedance changes. Ground plane asymmetry and measurement setup alignment further contribute to detuning. Such differences are normal in compact wideband MIMO antennas, and as long as both ports adequately cover the intended operating band with acceptable return loss, the shift is considered acceptable. Also, the isolation is more than 17.0 dB between the port of RP1 & port of RP2 and more than 25.0 dB between the port of RP2 & port of RP1 as recorded from Fig. 12g.

The radiation gain of the proposed MIMO antenna is plotted in Fig. 13a and b. The Figure shows that the antenna exhibits a gain of 1.68 dBi at a resonant frequency of 5.9 GHz. The gain increases gradually in the upper frequency band with a maximum value of 2.4 dBi at 7.1 GHz, coupled with the overall wideband response, thus validating its effective radiation with stable gain performance in the operating frequency range. Also, the radiation efficiencies for both ports are plotted in Fig. 13. It is observed that both plots overlap closely and exhibit an efficiency of 76% at 5.9 GHz. Overall, a steady increase in efficiency is observed over the entire operational frequency range, justifying a consistent and reliable radiation efficiency, validating the antenna’s effectiveness for automotive communication systems.

Fig. 13.

Far-field analysis of the proposed two port MIMO antenna (a) Gain (b) Radiation efficiency.

Diversity performance of the two-port MIMO system (ECC_V2X, DG_V2X, TARC_V2X, CCL_V2X, MEG_V2X)

To ensure the reliable performance of the designed MIMO antenna system in multipath signal environments with enhanced diversity gain, various parameters such as ECC_V2X, TARC_V2X, and CCL_V2X need to be evaluated. Further, a low value of TARC_V2X and ECC_V2X is desirable to signify low mutual coupling and low correlation between their respective radiation patterns, and thereby enhanced diversity gain. Also, CCL_V2X needs to be below its acceptable limit to support high data throughput. The performance of the proposed antenna based on Diversity parameters is presented in this section as detailed below:

The isolation and diversity performance between the MIMO system’s two closely positioned antenna elements is determined by calculating ECC_V2X using its S-parameters as detailed in^42–44.

9

10

11

For acceptable diversity performance, ECC_V2X should be < 0.5. The lower value of ECC_V2X is particularly desired to validate spatial diversity and minor correlation between ports. The plot representing ECC_V2X as a function of frequency (GHz) is shown in Fig. 14a, where simulated-measured results are compared. From the plot, the simulated results demonstrate a consistent decrease in the value of ECC_V2X from 0.15 at 4 GHz, dropping sharply to a value less than 0.001 beyond 5.3 GHz, and finally becoming negligible at higher frequency ranges. This near-zero value of ECC_V2X ensures that antenna elements are well isolated, owing to careful designing of the decoupling structure, spacing between elements, and their orthogonal placement. This validates the effectiveness of the proposed antenna for MIMO_V2X applications, justifying good isolation and better radiation behaviour. Also, the measured ECC_V2X is less than 0.05 in the V2X band.

Fig. 14.

Simulated and measured Diversity performance. (a) ECC_V2X (b)DG_V2X (c) TARC_V2X (d) CCL_V2X (e)MEG_V2X.

Diversity Gain (DG_V2X) evaluates the system’s ability to improve signal strength, particularly in multipath environments. Ideally, it must be close to 10 dB for a well-performing MIMO_V2X system, which indicates strong diversity performance and is calculated using the following equation^42–45.

12

Figure 14b shows the DG_V2X plot for the proposed MIMO_V2X antenna with respect to frequency ranging from 4 to 8 GHz. It is observed that the gain increases monotonically from 9.2 dB at 4 GHz to 9.99 dB at the desired centre frequency of 5.9 GHz and remains stable beyond this point.

Also, the figure shows that the proposed MIMO_V2X system exhibits consistently high diversity gain and low distortion over a wideband range, validating reliable signal strength characteristics suitable for V2X communication systems. The measures DG_V2X are greater than 9.95 dB in the V2X communication channel noted from Fig. 14b. To analyse the real reflection behaviour of the proposed two-port MIMO_V2X antenna, TARC_V2X is calculated and plotted in Fig. 14c. It effectively measures the ratio between reflected total power with respect to the incident total power when both the ports are excited simultaneously, and is governed by the following equation^42,45:

13

14

15

16

17

Here, the incident and the reflected wave at the i_th port are represented by ai and bi, respectively, where N denotes the total number of ports.

The plot shows that the TARC_V2X for the proposed MIMO_V2X antenna varies with the frequency and exhibits a value of < − 8.0 dB in the range of 5.13 GHz to 7.10 GHz for simulation results and less than − 4.0 dB for measured results.

Channel capacity, as defined in Shannon’s theorem, is the maximum rate at which information is transmitted over a communication channel without any interference and is a crucial parameter to determine the performance of an MIMO_V2X system. Thus, it quantifies the number of bits transmitted reliably per Hz of bandwidth⁴⁶. Channel capacity enables the assessment of the performance of an MIMO system with that of a single antenna. It largely depends on other diversity performance parameters and increases linearly with an increase in the antenna elements in the MIMO_V2X system configuration⁸. Further, the degradation in the system is evaluated using channel capacity loss (CCL_V2X) and calculated using the following equations:

18

19

where

20

21

The simulated-measured CCL_V2X shown in Fig. 14d are less than 0.05 b/s/Hz and 0.0025 b/s/Hz, respectively.

Mean Effective Gain is a crucial parameter in evaluating the performance of MIMO_V2X parameters as it estimates the antenna’s influence on the total link budget in a real propagation environment. It is defined as the ratio of the power received by the proposed MIMO_V2X antenna to the average power that would be received by two isotropic antennas placed under the same environment”⁴⁶. It is calculated using the determined S-parameters of the proposed MIMO_V2X antenna system, using the following equation:

22

23

24

The ratio calculates the MEG_60.0GHz, which is given by

25

where M is the number of antenna elements in the MIMO_V2X system, η_{i, rad} is the radiation efficiency for the i^th port, and S_ij are the S-parameters between the ith and jth ports, respectively. Figure 14e records the simulated-measured MEG between port1 and port2 (MEG₁₂), which is around − 3.0 dB, and the ratio of MEG₁/MEG₂ for both results is around 0.0 dB.

Experimental characterization and validation, state-of-the-art comparison

Figure 15 shows the detailed analysis of the MIMO antenna loaded with FSS, where simulated and measured far-field results are discussed. Figure 15a shows the MIMO antenna integrated with a 5 × 5 FSS designed for V2X communication. The MIMO antenna with a two-port configuration is placed above the FSS by an air gap distance of d_A-FSS = 25.0 mm, approximately at 5.90 GHz. The measurement of the MIMO-antenna integrated with FSS is shown in Fig. 15b, which is placed within the anechoic chamber for radiation-pattern, peak-realized-gain (PRG), and radiation-efficiency (RE) measurement at 5.90 GHz. Figure 15c and d show the 3D-radiation patterns at 5.90 GHz without and with FSS.

Fig. 15.

Far-Field analysis. (a), (b) Simulation model of MIMO antenna loaded with FSS (b) The measurement setup within anechoic chamber (c), (d) 3D radiation patterns at 5.90 GHz without and with FSS; Simulated-Measured. (e), (f) 2-D radiation patterns at 5.90 GHz without FSS in E–H plane (g) 2-D radiation patterns at 5.90 GHz in boresight direction. (h) Peak-realized-gain and radiation efficiency at 5.90 GHz.

Figure 15c reflects the 3D-radiation pattern at 5.90 GHz with dipole and omnidirectional patterns in principal planes (E–H). Figure 15e and f record the simulated and measured 2D-radiation patterns without loading of the FSS array in the E–H plane, both in co- and cross-polarizations. The simulated-measured radiation pattern in E-plane resembles a dipole-type pattern with minimal cross-polarisation. Also, the H-plane simulated and measured 2D radiation patterns look like omnidirectional patterns, which also record reduced cross-polarisation. Figure 15g shows the comparison of simulated-measured 2D radiation pattern in the boresight direction at 5.90 GHz. The simulated peak-gain with MIMO-antenna integrated FSS-array corresponds to 7.96 dBi with the main lobe directed at 27° and an angular 3.0 dB width of 35.1°. Also, the measured peak-gain records the value of 7.58 dBi with an angular 3.0 dB width of 38.4°. In both the above cases, the side lobe level value corresponds to − 2.60 dB, while the measured side-lobe is − 3.60 dB. These two values ensure a reduction in side-lobe, and the reflected signal from FSS placed below the MIMO antenna gets algebraically added with the main lobe, which enhances the peak-realized-gain.

Figure 15h shows the simulated-measured peak-realised-gain at 5.90 GHz without and with FSS. In the absence of FSS, the peak-gain at 5.90 GHZ records the value of 2.75 dBi. The peak-gain in the presence of FSS corresponds to 7.96 dBi in simulation and 7.58 dBi in the measured environment at 5.90 GHz, respectively. Figure 15h also records the radiation efficiency, which is more than 80% at 5.90 GHz, both in simulation and measured results.

In the proposed work, the enhancement of the gain is achieved by integration of reflector which is largely relative bigger in terms of physical aperture. The overall footprint of the antenna system increases but the reflector effectively improves the forward radiation and also front-to-back ratio. However, the larger size of the FSS do face challenge for practical integration in vehicular platforms where compact size, mechanical constraints and ease of installation are vital requirements. However, the compactness claim in this proposed work refers to the antenna size itself. The future work to achieve larger gain will focus on miniaturization of the unit FSS (reflector) cells, AMC/EBG-based reflectors or also integration of conformal techniques which will enhance suitability for real-time vehicular environments.

The Comparison of the proposed work with existing literature is given in the Table 4. The two antennas, with overall sizes of 55 × 79 mm²¹² and 19 × 43 mm²²⁸, enhance the peak-gain by 3.0 dBi and 5.84 dBi; however, the proposed antenna occupies a compact dimension of 24 × 24 mm² with enhanced peak-gain by 4.82 dBi. The remaining published work does not include the loading of FSS and records the nominal peak-gain around 5.0 dBi. Hence, the proposed antenna with gain enhancement and compact size with good diversity parameters outclasses the earlier published work and is highly recommended for V2X communication, which can be easily integrated with moving automobiles.

Table 4.

Performance comparison of the proposed antenna with recently developed MIMO antennas for V2X Communication Applications.

References	Size (mm2)	No. of Ports	Bandwidth (GHz)/Isolation (dB)	ECC DG (dB)	TARC (dB) CCL (b/s/Hz)	Loading of FSS (Y/N)	Gain enhancement
1 2022	30 × 76.66 0.177λ0 × 0.1475λ0	2	2.95–6.02  > 18.20	< 0.02  > 9.99	< -10.0  < 0.078	N	N
4 2024	80 × 80 0.603λ0 × 0.768λ0	4	5.84–6.01  > 17.0	< 0.028  > 9.998	< -3.0  < 0.35	N	N
6 2020	150 × 100 0.182λ0 × 0.133λ0	2	2.50–2.55  > 20.0	< 0.003  > 9.99	NC NC	N	N
7 2016	27.69 × 97.0 0.451λ0 × 0.842λ0	2	5.49–6.024  > 33.0	< 0.026 NC	NC NC	N	N
12 2019	55 × 79 0.594λ0 × 1.009λ0	2	5.42–5.95  > 18.0	NC NC	NC NC	Y	3.0
13 2020	45 × 55.6 1.5918λ0 × 2.6435λ0	2	5.685–5.884  > 7.0	< 0.02  > 9.952	NC NC	N	N
15 2024	52 × 77.50 0.456λ0 × 0.456λ0	2	3.03–3.44  > 18.20	< 0.01  > 9.998	< − 10.0 NC	Y	3.57
18 2020	37 × 44 2.29λ0 × 2.29λ0	2	5.22–5.52  > 23.0	< 0.10  > 9.94	NC  < 0.09	N	N
28 2024	19 × 43 0.626λ0 × 0.626λ0	2	3.0–6.0  > 10.0	< 0.004  > 9.99	< − 10.0  < 0.20	Y	5.84
47 2023	60 × 87 0.9798λ0 × 0.9798λ0	2	5.85–5.95  > 30.0	< 0.001  > 9.80	< − 10.0  < 0.32	N	N
34	50 × 50 0.44λ0 × 0.516λ0	2	3.0–6.0  > 20.0	< 0.10  > 9.90	< − 1.0  < 0.40	Y	4.80
35	30 × 30 0.358λ0 × 0.447λ0	1	3.30–10.80 NC	NC NC	NC NC	Y	8.10
37	50 × 50 0.370λ0 × 0.414λ0	4	3.14–12.24  > 20.0	< 0.004  > 9.90	< − 14.41 NC	N	N
Proposed	24 × 24 0.53λ0 × 0.53λ0	2	4.12–6.92  > 17.0	< 0.001  > 9.995	< − 4.0  < 0.0025	Y	4.82

ECC, Envelope correlation coefficient; DG, Diversity gain; TARC, Total active reflection coefficient; CCL, Channel capacity loss; FSS, Frequency selective surface.

In contrast to existing V2X MIMO antennas reported in the literature, which often rely on bulky decoupling structures or single-band optimization, the proposed antenna offers a compact radiator geometry capable of supporting multiple V2X-relevant bands with high isolation. The integration of the petal-shaped radiator with a connected DGS enables effective isolation enhancement while preserving radiation efficiency. Furthermore, the FSS reflector is specifically designed to operate synergistically with the MIMO configuration, providing gain improvement without degrading envelope correlation coefficient (ECC) or diversity performance.

Conclusions

A two-port MIMO antenna (24 × 24 × 0.8 mm³) loaded with frequency-selective-surface array (204 × 204 × 1.60 mm³) configuration is investigated. Petal-shaped radiating-patch with rectangular ground printed on opposite surface of FR4 dielectric generated measured operational-bandwidth of 4.80 GHz-6.82 GHz which is used for vehicle-to-everything (V2X) wireless communication. Also, 5 × 5 FSSV2X array with each element generating 2.0–9.0 GHz is used as reflector placed below MIMO antenna achieves maximum peak-realized-gain of 7.58 dBi at 5.90 GHz. The resonance frequency of V2X band is also validated by equivalent-circuit-model with the MIMO diversity parameters ECC_V2X < 0.001, DG_V2X > 9.950 dB, TARC_V2X < − 8.0 dB and CCL_V2X < 0.05 b/s/Hz. Thus, the designed MIMO antenna, being compact, with a low profile, exhibiting high isolation, justifies itself to be a potential candidate for next-generation intelligent transportation systems. The novelty of this work does not stem from introducing entirely new antenna elements, but from the effective combination and optimization of known techniques to address multiple performance metrics simultaneously for V2X MIMO applications. The proposed design demonstrates how these techniques can be jointly employed to achieve multi-band operation, improved isolation, and enhanced gain within a single antenna framework.

The proposed work is explicitly clarified in the manuscript that system-level validation is beyond the current scope and identified it as a future research direction. Future work will integrate the proposed antenna into a complete V2X transceiver framework to evaluate link budget margins, BER performance under Doppler effects, and performance over standardized vehicular channel models, thereby enabling end-to-end validation.

Author contributions

Bhaskara Rao Perli, K Sathish, Aarti Bansal—worked on Conceptualisation and methodology; Tathababu Addepalli—Conducted the formal analysis. B. Satya Sridevi—conducted the experiment, investigation and review the original draft. Manish Sharma—Review. Kanhaiya Sharma—Review.

Funding

Open access funding provided by Symbiosis International (Deemed University).

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Competing interests

The authors declare no competing interests.