Antenna configuration

In the design presented in this work, the configuration of printed dipole antennas with integrated balun feeding is used. These antennas have wide bandwidth and stable and acceptable radiation behavior [Reference Shokri, Ghobadi and Nourinia22–Reference Gou, Yang, Li and Nie25]. The antenna configuration proposed in this study is demonstrated in Fig. 1.

Figure 1. (a) 3D view of the proposed antenna, (b) dipole arms, (c) parasitic elements, and (d) balun structures. (All dimensions are in millimeters).

Slotted diamond-shaped dipole arms and arrow-shaped parasitic elements are used in this design. This figure shows an FR4 substrate with a relative permittivity of 4.4, a loss tangent of 0.02, and a thickness of 1 mm with dipole arms and parasitic elements printed on the bottom and top layers, respectively. This substrate is located at the top of the antenna. It is possible to achieve the cross-shaped arrangement of the baluns related to the dipoles by creating slots above and below the two FR4 substrates with a thickness of 0.8 mm in the middle part of the antenna. Baluns are soldered to the dipole arms from the top and the ground plane from the bottom. Also, an FR4 substrate with a thickness of 0.8 mm is located in the lower portion of the antenna, which supports other parts mechanically and establishes the ground plane on both sides. The input signals are transmitted to the antenna through two coaxial cables soldered from the bottom of the lower substrate to the antenna. Thus, the external conductors of the coaxial cables are soldered to the ground plane of the substrate, and the internal conductors are soldered to the Γ-shaped feeding of the baluns after passing through two holes in the substrate. The baluns proposed in this paper consist of a Γ-shaped feed on one side of the substrate, and also, there are two shorted patches that serve as a slot line on the other side of the substrate. The input signal to the Γ-feed is coupled from the horizontal part to the slot line, where it reaches the dipole arms. Consequently, dipoles-1 and -2 are placed orthogonally on each other, and because dipoles are linearly polarized, they produce horizontal and vertical linear polarization. On the other hand, by rotating the antennas by 45°, the ±45° slant polarizations are formed.

Designing a printed dipole antenna with integrated balun for BTS applications in the 1.7–2.7 GHz frequency range involves several equations and calculations. The dimensions of the antenna segments are critical to ensure proper operation within the specified frequency range. In the first step, the center frequency of the design should be determined. The proposed antenna in this work should cover the frequency range 1.7–2.7 GHz for BTS applications. The central frequency (f ₀) can be taken as the average of this range:

(1)\begin{equation}{f_0} = \frac{{{f_l} + {f_h}}}{2} = \frac{{1.7 + 2.7}}{2} = 2.2\,GHz\end{equation}

So, the wavelength (λ) corresponding to the central frequency (f ₀) is equal:

(2)\begin{equation}\lambda = \frac{c}{{{f_0}}} = \frac{{3 \times {{10}^8}}}{{2.2 \times {{10}^9}}} \approx 136\,mm\end{equation}

where c is the speed of light. Consequently, for a half-wave dipole, each arm length is approximately:

(3)\begin{equation}L = \frac{\lambda }{4} = \frac{{136}}{4} = 34\,mm\end{equation}

However, due to the effects of the dielectric substrate and the actual operational environment, the effective length might be slightly shorter. Typically, corrections are made for the substrate’s effective dielectric constant (ε_r). In the presented design, the diamond-shaped dipole arms are connected from the middle to the short-circuited patches of the balloon, so their length is proportional to the provided equations. However, parametric studies have also been conducted to obtain the optimal parameters.

In the second step, the dimension of the balun should be calculated. The balun is designed to convert the unbalanced coaxial feed line to a balanced antenna. The balun can be integrated with the dipole on the same substrate. Commonly used substrates for printed antennas include FR4, Rogers, and other materials with known dielectric constants (ε_r). The choice of substrate affects the dimensions due to the effective wavelength in the substrate. For microstrip lines, the effective dielectric constant influences the effective wavelength (λ_eff). Assuming a dielectric substrate with ε_r = 4.4 (for example FR4):

(4)\begin{equation}{\lambda _{eff}} = \frac{\lambda }{{\sqrt {{\varepsilon _{eff}}} }}\end{equation}

where:

(5)\begin{equation}{\varepsilon _{eff}} = \frac{{{\varepsilon _r} + 1}}{2} = \frac{{4.4 + 1}}{2} = 2.7\end{equation}

Therefore:

(6)\begin{equation}{\lambda _{eff}} = \frac{{136}}{{\sqrt {2.7} }} = 83\,mm\end{equation}

Consequently, the effective length of the dipole arms should be equal to:

(7)\begin{equation}{L_{eff}} = \frac{{{\lambda _{eff}}}}{4} = \frac{{83}}{4} = 21\,mm\end{equation}

These calculations provide a starting point. The optimal dimensions should be reconsidered through simulation and parametric studies to account for additional factors, such as nearby structures (parasitic elements in this work), precise substrate characteristics, and feed structure (including balun and connector).

Antenna without parasitic elements

At first, the aim was to design a crossed-dipole antenna capable of producing dual polarization. The dipole arms were initially configured in a diamond-shaped formation, as shown in Fig. 2(a). As a result of the parametric analysis, it has been decided to modify this shape into a slotted diamond-shape. According to the investigation, the modifications made to the dipole arms have enabled a suitable bandwidth for the antenna. This results in a frequency bandwidth of 1.48–3.93 GHz and 1.57–4.24 GHz, respectively, for the crossed-dipole antennas in ports -1 and -2. Figure 2(b) illustrates the results of the simulations performed on the antenna with both structures introduced for the dipole arms.

Figure 2. (a) Two configurations for dipole arms (#1 and #2) and (b) S-parameters of the antenna for configurations of #1 and #2 for dipole arms.

Additionally, it is crucial to investigate the configuration of the Γ-feed. A detailed analysis of this issue can be found in Fig. 3.

Figure 3. Two configurations for Γ-feed (#uniform Γ-feed and #proposed Γ-feed), in (a) dipole-1, (b) dipole-2, and (c) S-parameters of the antenna for two configurations for Γ-feed. (All dimensions are in millimeters).

Based on this figure, two modes of the Γ-feeding can be considered, the uniform mode and the modified mode (proposed). Simulated results demonstrate the superiority of the modified mode over the uniform mode, and the IBW is obtained as before. The bandwidth has a criterion of |S ₁₁| or |S ₂₂| < −10 dB, which does not meet the requirements of BTS systems. As a result, a further step needs to be taken in antenna design to increase the IBW sufficiently. According to Figs. 1(d) and 3, the dimensions of two Γ-feeds have been chosen differently in order to prevent a connection between them and the possibility of them passing over one another, but the difference in sizes has been taken into account in such a way that impedance matching and operational bandwidth to the results obtained in each antenna port are similar.

Theoretical foundations and physics of antenna design

An antenna’s IBW is defined as the frequency range over which the reflection coefficient |S ₁₁| is less than a certain threshold, commonly −10 dB. However, the threshold is more stringent at −15 dB in this case. Generally, the reflection coefficient |S ₁₁| is related to the input impedance Z _in of the antenna as follows:

(8)\begin{equation}\left| {{S_{11}}} \right| = \left| {\frac{{{Z_{in}} - {Z_0}}}{{{Z_{in}} + {Z_0}}}} \right|\end{equation}

In this case, Z ₀ represents the characteristic impedance of the feed line, which is usually 50 Ω. Z _in must remain close to Z ₀ over a wide frequency range, minimizing |S ₁₁| and thus enhancing IBW. The parasitic elements are passive structures not directly fed from the antenna’s feed line but are in contact with the driven elements (the dipoles) via mutual coupling. Their effects can significantly impact the antenna’s input impedance and radiation characteristics. The resonant frequency f _r of a dipole antenna is determined by its length L and the effective wavelength λ_eff:

(9)\begin{equation}{f_r} \approx \frac{c}{{2L\sqrt {{\varepsilon _{eff}}} }}\end{equation}

where c is the speed of light in free space, and ε_eff is the effective dielectric constant of the substrate. The presence of parasitic elements near dipoles leads to additional resonances due to their own geometry and coupling with the dipoles. Depending on their placement and dimensions, these parasitic elements can either enhance or suppress specific resonant modes. The mutual coupling between a driven dipole and a parasitic element can be analyzed using the mutual impedance Z ₁₂:

(10)\begin{equation}{Z_{12}} = \int_{{V_1}} {\int_{{V_2}} {{J_1} \cdot } } {E_2}\,d{V_1}\,d{V_2}\end{equation}

where J ₁ is the current density on the driven element, and E ₂ is the electric field produced by the parasitic element. The formula provided for mutual impedance is a general expression used to describe the interaction between two antennas (or antenna elements), such as a driven dipole and a parasitic element (in this case). In the context of this formula, V ₁ denotes the volume over which the current density J ₁ is distributed on the driven element (the dipole arms in this case). The volume V ₁ covers the region occupied by the driven dipole, including any significant portions of the near-field where the electromagnetic fields generated by the dipole are strong. Similarly, V ₂ represents the volume over which the parasitic element’s electric field E ₂ is considered. This volume encompasses the region occupied by the parasitic element, including the areas characterized by electromagnetic fields. Furthermore, J ₁ represents the current density on the driven element (dipole). It is a vector quantity describing the distribution and magnitude of the current flowing on the surface of the driven element, and E ₂ refers to the electric field generated by the parasitic element. This field is a result of the currents induced in the parasitic element by the electromagnetic fields originating from the driven element (dipole). Furthermore, dV ₁ and dV ₂ represent differential volume elements within the volumes V ₁ and V ₂. The integrals sum up the contributions of the mutual coupling over the entire volumes where the fields and currents are present.

On the other hand, the mutual impedance Z ₁₂ quantifies the interaction between the two elements (driven and parasitic elements) by integrating the product of the current density on the driven element and the electric field generated by the parasitic element over the volumes V ₁ and V ₂. As a result of this interaction, the overall impedance of the driven dipole is affected, which in turn affects the performance of the antenna, such as its IBW and radiation pattern. It is essential to understand and calculate the mutual impedance Z ₁₂ in this design. It enables the determination of how parasitic elements affect the input impedance of the driven dipole, and by extension, the antenna’s operating characteristics. The total input impedance Z _in of the antenna becomes:

(11)\begin{equation}{Z_{in}} = {Z_d} + {Z_{12}}\end{equation}

where Z _d is the self-impedance of the driven dipole. The parasitic elements, when optimally designed and positioned, can make additional resonant paths and increase the total adequate bandwidth of the antenna. This is because they allow the antenna system to support multiple resonant modes at slightly different frequencies, which combine to achieve a broader overall resonance. Based on the derivative of the impedance concerning frequency, it is possible to estimate the total IBW:

(12)\begin{equation}IBW \propto {\left( {\frac{{d\left| {{Z_{in}}} \right|}}{{df}}} \right)^{ - 1}}\end{equation}

Z _in is smoothed out over a wider frequency range by the introduction of parasitic elements, thereby reducing the sharp variations that would otherwise limit its bandwidth. Parasitic elements can also be used to reduce the physical size of an antenna without compromising its performance. As a result of carefully choosing the shape and location of parasitic elements (in this case, arrow-shaped), the effective electrical length of the dipoles can be increased, allowing the antenna to resonate at a lower frequency than a simple dipole of the same physical dimensions. This is mathematically represented by an effective length L _eff of the dipole, which is extended by the parasitic elements:

(13)\begin{equation}{L_{eff}} = L + \Delta {L_p}\end{equation}

where ΔL_p is the contribution from the parasitic elements. Consequently, the antenna can maintain its resonant frequency at a given size or achieve the same frequency with a reduced size:

(14)\begin{equation}{f_r} \propto \frac{1}{{{L_{eff}}}}\end{equation}

Parasitic elements also affect the radiation pattern and gain. The parasitic elements can create constructive and destructive interference patterns, which can affect the total radiated power Prad and the directivity D of the antenna. An antenna’s gain G can be calculated as follows:

(15)\begin{equation}G = D \times \eta \end{equation}

where η represents the antenna’s efficiency. Parasitic elements can modify the current distribution on dipoles, increasing directivity and thus increasing gain in certain directions. Simulations (e.g., using High-Frequency Structure Simulator (HFSS) or Computer Simulation Technology(CST)) and experimental measurements are often used to validate the effects of parasitic elements on impedance and bandwidth. Typically, the design process involves iterating the geometry and placement of parasitic elements in order to achieve the desired IBW, gain, and overall performance. By introducing additional resonant modes and smoothing the impedance variation over frequency, parasitic elements increase the IBW. As a result, the antenna’s physical dimensions can be reduced by increasing the electrical length of the dipoles. Parasitic elements can enhance the gain and shape the radiation pattern by altering the current distribution and mutual coupling effects.

Proposed antenna, results, and discussion

Examining the concept of magnetic coupling between the dipole arms and the parasitic elements is necessary to understand how the parasitic elements work. In this section, the performance of the proposed antenna is investigated by employing parasitic elements. Magnetic coupling in a dipole antenna can improve its bandwidth and reduce its dimensions. This technique is known as “mutual coupling” or “inductive coupling.” The magnetic field generated by the current flowing through the dipole induces currents in parasitic elements when placed near the dipole arms. If certain conditions arise, the induced currents will interact with the currents in the dipole arms constructively, resulting in a broader frequency response, expanding the IBW of the antenna and allowing it to operate over a more comprehensive range of frequencies. Consequently, the antenna’s effective length can be increased by employing parasitic elements without physically extending the dipole arms. Therefore, it allows for a more compact antenna design while maintaining resonant characteristics. It is concluded that the interaction between the parasitic elements and dipole arms effectively extends the antenna’s electrical length, reducing the dipole’s physical dimensions and making the antenna more compact. Figure 4 compares the antenna’s IBW with and without parasitic elements.

Figure 4. S-parameters of the proposed antenna with/without parasitic elements.

Concerning the figure, it is clear that the presence of parasitic elements next to the dipole arms significantly increases the IBW of the antenna with the criterion of |S ₁₁| or |S₂₂| < −15 dB. Meanwhile, parasitic elements are necessary for the proposed antenna to satisfy this criterion regarding IBW. Based on the simulation results, it can be found that utilizing parasitic elements in the proposed antenna structure makes the antenna can cover the frequency ranges of 1.53–3.03 GHz and 1.55–2.93 GHz for ports -1 and -2, respectively, with a criterion of |S ₁₁| or |S ₂₂| < −15 dB. It is also essential in antenna theory to study the distribution of surface currents on dipole arms and parasitic elements. This analysis is necessary to understand an antenna system’s behavior and performance. The surface current distribution on the proposed antenna with and without parasitic elements is depicted in Fig. 5. It can be seen in the figure that the accumulation of surface currents on the dipoles in which parasitic elements are not present is more accumulated in the slot sections. In addition, when parasitic elements are added to an antenna structure, some of the surface currents are also generated on the parasitic elements, increasing antenna IBW. To better understand this phenomenon, we also analyzed the impedance of the proposed antenna.

Figure 5. Current distribution on the (a) proposed antenna without parasitic elements and (b) with parasitic elements.

The input impedance of an antenna is a complex quantity with real and imaginary parts that characterize the opposition to the current flow in an electrical circuit model. In the case of a dipole antenna, the impedance consists of a resistive component (real part) and a reactive component (imaginary part). Indeed, the power loss is represented by the resistive component of the impedance and is denoted by the real part of the impedance, typically measured in ohms. Several factors contribute to the real part, including the resistance of the conductive elements, losses in the antenna structure, and radiation resistance. Instead, the impedance’s reactive component represents the energy storage and release in the antenna. Similarly, it is measured in ohms and represents the imaginary part of the impedance. Depending on the frequency of operation and the physical characteristics of the antenna, the reactance can either be capacitive or inductive. As a result, a dipole antenna’s overall impedance is determined by the cooperation of resistance and reactance, which is essential to have a thorough understanding of both components to optimize the dipole antenna’s performance. Considering the physical dimensions, spacing, and electrical properties of the parasitic element, its effect on the impedance of the dipole can be taken into account. In addition, analytical methods using electromagnetic field analysis techniques can provide valuable insights into the impedance behavior of the parasitic element. Parasitic elements can be analyzed using electromagnetic simulation software such as the HFSS package. Using numerical methods, the software tools solve Maxwell’s equations and simulate the behavior of electromagnetic fields in the antenna system. The impedance values of the proposed antenna with and without parasitic elements are plotted in Fig. 6 regarding real and imaginary parts.

Figure 6. The impedance (real and imaginary parts) of the proposed antenna with/without parasitic elements.

The simulation results show that the real and imaginary parts of the antenna input impedance in the presence of parasitic elements are more inclined to the values of 50Ω and 0Ω, respectively, and show better impedance matching.

Antenna fabrication and measurement results

Antennas in base stations may require a wide bandwidth in modern mobile communication systems. For example, 2G/3G/4G/LTE systems operating in the 2 GHz band must cover the frequency range of 1.71–2.69 GHz, that is, 2G (1710–1920 MHz), 3G (1880–2170 MHz), and 4G/LTE (2300–2400 MHz and 2570–2690 MHz). As illustrated in Fig. 7(a), a comparison between the measured and simulated IBWs of the proposed BTS antenna is presented. Experimental results show this design can cover 1.72–2.71 GHz and 1.58–2.72 GHz frequency bands with |S ₁₁| or |S ₂₂| < −15 dB criterion for ports-1 and -2, respectively.

Figure 7. Simulated and measured S-parameters of the proposed antenna, (a) impedance bandwidth, and (b) isolation.

Furthermore, as reported in Fig. 7(b), the simulated and measured isolations (−|S ₁₂| and −|S ₂₁| in dB) are higher than 17 dB in the operating frequency band. The proposed antenna employs a crossed-dipole configuration, which inherently has dual polarization. This design allows the antenna to exhibit ±45° slant polarizations (or vertical/horizontal linear polarizations as an alternative) by positioning the two orthogonally placed dipoles at ±45° to the horizontal plane (or 0° and 90°).

These slant polarizations are crucial in BTS applications as they minimize polarization mismatch and enhance signal reception in environments with multipath propagation. Furthermore, when aligned with the vertical and horizontal axes, the antenna can also function with V/H linear polarizations, providing versatility in various operational scenarios. We have confirmed the dual-polarization performance through experimental measurements, demonstrating excellent isolation between the two polarizations. In addition, the proposed antenna has a measured peak gain of 7.8 dB in the operating frequency band. Figure 8 shows the simulated and measured results of the proposed BTS antenna gain in ports-1 and -2.

Figure 8. Simulated and measured realized gain of the antenna.

In this work, all the simulated results were extracted using HFSS Version 15 software, and the experiments were carried out using an Agilent E8363C network analyzer. The numerical and experimental results show a good agreement, and, in some frequencies, there is a slight discrepancy between the numerical and experimental results due to errors made during antenna fabrication and testing. Simulated and measured normalized H- and E-plane radiation patterns of the proposed antenna for ports-1 and -2 are plotted in Fig. 9.

Figure 9. Simulated and measured normalized H- and E-plane radiation patterns of the proposed antenna for (a) port-1 and (b) port-2.

The patterns are extracted at two frequencies of 1800 and 2200 MHz based on co- and cross-poles. According to the results, the proposed BTS antenna has measured half-power beamwidths (HPBWs) of 62.15° ± 1.45° at two sample frequencies. Furthermore, using a smooth ground plane under the antenna makes its radiation pattern unidirectional at both planes. Following the design and optimization of the proposed BTS antenna, it was fabricated using FR4 substrates and tested in the antenna laboratory. Photographs of the antenna prototype made in this work are shown in Fig. 10. This figure also depicts how to test the antenna’s S-parameters using a network analyzer and its radiation parameters using an anechoic chamber.

Figure 10. Photographs of the (a) manufactured prototype, (b) antenna connected to the Vector Network Analyzers (VNA), and (c) antenna at an anechoic chamber.

The proposed BTS antenna offers significant advantages and is superior to similar works regarding bandwidth and dimensions. In addition, due to its smaller dimensions, it is more flexible in installation, making it suitable for various environments and reducing the visual impact. With increased bandwidth and compact size, BTS installations are not only able to optimize network performance but also enhance their aesthetic appeal. The results of this comparison are tabulated in Table 1.

Table 1. Comparison of antenna measured results

The proposed theory in this work on printed dipole antennas with integrated baluns and Γ-shaped feed elements could be extended in future research. The authors appreciate the reviewers’ suggestion to explore our proposed model’s future potential and extensions. Several avenues for further research could enhance the impact and relevance of our work on printed dipole antennas with integrated baluns and Γ-shaped feed elements.

Fundamentally, future work could explore adapting our antenna design to support wideband or multiband operation. By optimizing the geometric parameters and incorporating additional elements, the antenna could be made to operate efficiently across multiple frequency bands, such as those used in modern communication systems (e.g., 5G, Wi-Fi, IoT). Furthermore, scaling the design for higher frequency bands, such as millimeter-wave frequencies (24–100 GHz), could be investigated. This technique addresses challenges related to increased losses and manufacturing precision at higher frequencies. Also, integrating active components, such as high frequency diodes, varactors, or MEMS switches, can lead to reconfigurable antennas capable of dynamically adjusting their frequency, polarization, radiation pattern, and gain. This could provide greater flexibility and adaptability for various applications.