1 Introduction

Over recent years, automotive technology has been rapidly changing, which requires a significant increase in the data rate transferred. Development in the wireless network industry is becoming very necessary to meet the huge demand for such data rates. Many countries have adopted the fourth generation (4G) wireless communication system. However, some drawbacks, including overcrowded spectrum, high energy consumption, and high data rates, are not solved by 4G [1]. In comparison to the previous generation of 4G, 5G introduces significant differences and promises more benefits, such as wider bandwidth and a less congested spectrum. The main goal of 5G technology is to improve the data rate by 1000 times compared to common communication standards and to connect 100 billion devices. The high data rates of several Gbps need even more support from the use of higher frequency bands, so 5G needs a wide spectrum across low, mid, and high ranges to deliver widespread coverage and support all use cases. AutomotiveWireless communications include various applications, including automated driving of vehicles, driver assistance systems, vehicle-to-everything (V2X), vehicle-to-vehicle (V2V), and vehicle-to-infrastructure (V2I) [2]. The automotive applications need antennas that have some properties like beam steerable, wide bandwidth, and high realized gain. However, the automotive applications are loaded with optional antennas, so the proposed system is loaded with an antenna that has beam-steering, high gain, and wide bandwidth properties to work in different environmental conditions to cover important automotive services and support a high data transfer rate [3]. Antenna-array systems with butler-matrix are one of the methods to achieve the desired aims. In this paper, the single-element antenna is a Vivaldi, which was suggested by Gibson [4]. The single element is loaded with NZIM to improve the overall realized gain [5]. Beam steering is a technique to steer the beam by controlling the phase of each element. The [4 × 4] butler matrix is optimized for this aim [6, 7].

2 Single element design

A single Vivaldi element is designed and optimized to operate at 3.6 GHz with broadband characteristics. The Vivaldi antenna meets the desired purpose [8]. Roger RT5880 with relative permittivity (εr = 2.2), a thickness of 0.787 mm, and low tangent losses (tan δ = 0.0009) is chosen for sub-6 antennas to reduce losses, especially in antennas that operate at high frequencies. The single Vivaldi element geometry is shown in Fig. 1. The proposed design dimensions are presented in Table 1. Also, the simulated return loss |S11| is shown in Fig. 2. CSTMWS 2020 and HFSS, two software packages that differ in their simulation techniques, were used to simulate the proposed designed system [9].

Fig. 1
figure 1

The geometry of the proposed vivaldi antenna: a top view and b bottom view

Full size image
Table 1 The optimized dimensions of proposed vivaldi antenna
Full size table
Fig. 2
figure 2

The simulated and measured return loss S11 for the proposed vivaldi antenna

Full size image

The results simulated by the two programs are close to the measured results, achieving a − 10 dB bandwidth of 2.9 GHz for a single Vivaldi element.

3 Near-zero-index metamaterial [NZIM]

The phase velocity of propagating electromagnetic waves tends to infinity in the NZIM metamaterials. As a result, the final wavelength also tends to infinity. The NZIM metamaterial is used as a high refractive index impact to result in a confined pattern and high directivity [10]. Also, by providing a high-refractive-index effect, the metamaterial is intended to replace dielectric lenses for gain enhancement [10]. So, the NZIM units use the same thickness of dielectric material to be introduced within the front of a Vivaldi antenna [11, 12]. The NZIM metamaterial is used in essential applications such as producing unwanted impedance mismatch in the desired system and increasing the realized gain [13]. Based on the standard retrieval method, the CSTMWS 2020 is used to determine the effective relative permittivity (εr), relative permeability (μr), and refractive index (n) of the proposed NZIM cell [14] as shown in Fig. 3. This plot demonstrates that the proposed NZIM cell is operating quite effectively in the frequency band of 3–4 GHz, where the effective relative permittivity is close to zero. The NZIM unit cell dimensions are (Ls = 7 mm, Lm = 6 mm, and Wx = 0.5 mm) as shown in Fig. 3. The field vectors are connected by constitutive relations when electromagnetic fields are present in the material medium. When an electric field (E) passes through a dielectric material, the atoms or molecules polarize, producing electric dipole moments that increase the total displacement flux (\(\overline{{\varvec{D}} }\)) [15]. However, the NZIM unit cells are optimized such that the return loss |S11| is not significantly changed after their insertion into the front of a Vivaldi antenna [16] as shown in Fig. 2.

$$\overline{D} = \varepsilon r\, \varepsilon o \left( {1 + {\chi e}} \right)\overline{E}$$
(1)

where the electric susceptibility (χe) may be a complex.

Fig. 3
figure 3

a The geometry of proposed NZIM unit cell. b The real part of εr (relative permittivity), μr (relative permeability), and n (refractive index). c Image part of εr, μr, and refractive index (n)

Full size image

The value of the n index is approximately near zero, its real value is 0.1885, and the image value is approximately zero. The designed element of the Vivaldi antenna aperture is filled with NZIM to improve the overall realized gain, as shown in Fig. 4.

Fig. 4
figure 4

a The proposed designed antenna with embedded NZIM meta-material, b electric field distribution when the two columns of NZIM unit cells were added

Full size image

The near field phase fronts have been modified and produce some focusing effects, so that due to the NZIM cells, a plane wave approximation may be assumed rather close to the antenna than would be the case otherwise[15], as shown in Fig. 8(b). It was required to analyze the impact of the NZIM units on the realized gain after they were added to the proposed single element. Hence, the realized gain versus frequency is simulated using CSTMWS 2020 in Fig. 5.

Fig. 5
figure 5

The maximum realized gain versus frequency of a single Vivaldi element for various numbers of columns introduced within the front of the Vivaldi antenna element

Full size image

The value of the maximum overall realized gain is 8.6418 dB at a frequency of 3.6 GHz when three columns of the NZIM unit cells are added to a single Vivaldi element. The NZIM unit cells achieve a confined pattern with high directivity and increase the realized gain by 1.12 dB. The NZIM unit cells have a significant effect on the realized gain.

4 Butler matrix components design

Beam-steering is one of the important properties in automotive communications. The proposed system is fed by a Butler matrix, which is considered a passive beam-steering network. The Butler matrix was proposed by Butler and Lowe in 1961. It consists of a 90° hybrid Coupler, crossover, and phase shifter [17].

4.1 90° Hybrid Coupler

A hybrid coupler is a passive four-port network, as shown in Fig. 6, that splits the input power into two equal-power signals at the outputs, and the fourth port is isolated. The coupling is 3 dB and half-power split at the outputs. For a 90° hybrid coupler, the outputs differ in phase by 90° [6, 17].

Fig. 6
figure 6

The layout of the 3 dB coupler

Full size image

The 3-dB coupler is designed in HFSS and CSTMWS 2020 to obtain the S-parameters, and phase difference between coupled ports as shown in Fig. 7.

Fig. 7
figure 7

a The reflection coefficient of 3 dB coupler. b The phase difference between output ports, which is 90.4° at an operating frequency of 3.6 GHz

Full size image

4.2 Crossover

The crossover is optimized with a high level of isolation between the input ports to avoid overlapping signals during crossings. It is simple to construct by cascading two-hybrid 3-dB couplers [18]. The designed crossover and simulated results with good isolation between ports are shown in Fig. 8.

Fig. 8
figure 8

a The Layout of the crossover and b the reflection coefficient

Full size image

4.3 Phase shifter

In the designed system, phase shifters are critical components. The power at the output of the micro-stripline phase shifter should be identical to the input power, implying a 0 dB insertion loss. However, there are dielectric and conductor losses that degrade the insertion loss owing to the non-ideal transmission line length [19]. At higher frequencies, these losses are more pronounced. The phase difference between the input and output ports can be easily changed by changing the electrical length of the line [7].

4.4 The [4 × 4] Butler matrix

The butler matrix is designed, optimized, and fabricated. The simulation results are obtained using CSTMWS 2020 and HFSS. The output ports are separated by (λo/2), which is equal to approximately 42 mm at an operating frequency of 3.6 GHz to avoid mutual coupling between array elements. The proposed design layout is shown in Fig. 9.

Fig. 9
figure 9

The designed layout of the butler matrix

Full size image

The phase difference is estimated when firing the butler matrix input ports p1, p2, p3, or p4 and the output refrence port is p8. Also, when one of the input ports is firing, all of the input ports should be terminated with a 50 Ω. The results for the [4 × 4] butler matrix phase difference between output ports simulated by using CSTMWS 2020 and HFSS are shown in Fig. 10. When the input port P1 is firing and the output port P8 is the reference to the output ports .

Fig. 10
figure 10

The phase difference between output ports with respect to port 8 (p8). a In this case phase differences equal 45°, b 90°, and c 135° at the frequency of 3.6 GHz

Full size image

5 System design and result for discussion

In this section, the complete designed system results are obtained by using the Advanced Design System (ADS 2019) simulator to show mutual coupling between input ports and return loss for input ports [20]. The results were presented in Fig. 12 and compared with the measured ones. The data is extracted from CSTMWS 2020 (Touchstone files) and is used to obtain results in ADS 2019 as shown in Fig. 11.

Fig. 11
figure 11

The complete designed system data diagram in the ADS 2019 simulator

Full size image

The extrcted S-parameter data (Touchstone format) is imported to a butler matrix block using a S8P and an antenna element using an S1P. The simulation is run by finding the S-parameter matrix's biggest eigenvalue at each frequency point. The Butler matrix input ports are terminated with a 50 Ω.

Fig. 12
figure 12

a, b, c, and d are the mutual coupling between input ports of the designed system, which is less than − 20 dB. e and f is the return loss (RL) of the designed system

Full size image

The results obtained by ADS 2019 are approximately as measured. The |S11| is similar to the |S44|, and the |S22| is similar to the |S44|. To achieve beam steering, the phase of the input signal is adjusted for all radiating components [21, 22]. As shown in Table 2, the Butler matrix achieves this goal. The isolation between butler matrix input ports is denoted by the return loss (RL) |S12|, |S14|, |S24|, and |S23|. It can be figured out that the port isolation is very good as it achieves less than − 20 dB for all the ports over the operating band. Also, Fig. 13 shows beam steering simulated results by using CSTMWS 2020 compared to measured results.

Table 2 The phase difference between ports by using CSTMWS 2020 at an operating frequency of 3.6 GHz
Full size table
Fig. 13
figure 13

The realized gain versus theta in different cases by changing the input port, which leads to beam steering in cases (a), (b), (c), and (d)

Full size image

At the operating frequency of 3.6 GHz, the maximum measured realized gain is 11.2 dB, and the butler matrix enabled the system to cover an angle from 45° to 135°, which steered the beam in the desired direction as shown in Fig. 13. Also, the proposed system achieves a − 10 dB bandwidth of 600 MHz from |S11| and |S22| as shown in Fig. 12. The high gain, wide bandwidth, and beam steering technology in the designed system make communications possible in remote locations and critical environmental conditions for automotive applications. Beam steering in automotive applications helps to reduce multipath fading occurrences by forming the beam in the desired direction to avoid interference and noise sources. In addition, steerable antennas enable sensors to achieve longer ranges, higher data rates, and lower power consumption (longer battery life) [23, 24].

6 Conclusion

In this paper, the proposed design is simulated and fabricated, showing good results. The proposed single Vivaldi element supports the entire sub-6 GHz frequency bands and is loaded by the NZIM units to increase overall realized gain. The passive beam steering network is a butler matrix, which is optimized at an operating frequency of 3.6 GHz with good isolation between input ports, which is lower than −20 dB and covers an angle from 45° to 135°. The proposed system has some properties that support automotive applications, such as high gain, wide broadband, and the property of beam steering. Automotive applications are becoming increasingly important in our life, and include very essential services such as v2x, v2v, and global positioning system (GPS), so the proposed system can be considered as a good candidate for this purpose.