Antenna design geometry

The antenna geometry and dimensions are displayed in Fig. 1. A Rogers RT Duroid 5880 substrate spanning 33 mm × 30 mm × 1.6 mm and having ε_r = 2.2 and tan δ = 0.0009 is used to print the antenna. Relative permittivity of a substrate is an intrinsic property of the material. The dielectric constant of the substrates used for planar antennas varies in the range between 2.2 ≤ ε_r ≤ 12 [Reference Balanis27]. As we go up in the range the losses in the antenna increase, whereas when going down the range the losses get minimized and the antenna achieves maximum radiation capability. This maximized radiation gives a relatively good directive gain to the antenna element. The proposed antenna is designed on Rogers RT/Duroid 5880 which has a relative permittivity of (ε_r = 2.2) having low loss. Hence, the realized gain of the proposed antenna gets enhanced and reached to a peak value of 5.07 dBi. The antenna is composed of a partial ground plane with a rectangular shape and a patch with a hexagonal shape. Two 6.5 mm wide triangular extensions (L _g, W _g) are added to the ground plane. From the top edge of the ground plane, a single, rectangular slit approximately 4.5 mm (L _f, W _f) is etched out. By attaching the patch to the feedline, a thin 0.3 mm-wide strip is utilized to achieve reconfigurable characteristics. A 4.5 mm wide, 50 Ω microstrip line is used to feed the antenna, and it is optimized for reconfiguration of polarization. Polarization status and frequency can be adjusted by connecting those smaller rectangular metallic strips individually to the monopole as PIN diodes. This is because the adjustment alters the surface current distribution. The ground initially having a rectangular structure is further modified by adding two triangular-shaped metallic extensions and these are also attached to the ground as shown in Fig. 1(c). On the inside of the hexagonal patch, Koch curve-based fractal geometry is used to increase the bandwidth and provide appropriate impedance matching.

Figure 1.

Design overview of reconfigurable fractal antenna. (W _a = 30, L _a = 33, H _a = 1.6, L _b = 11.5, L _c = 7.5, W _d = 4.5, L _d = 14.5, W _e = 30, L _e = 12, W _g = 6.5, L _g = 6.5, W _f = 4.5, L _f = 4.5, all dimensions are in millimeters).

Reconfigurability is achieved by inserting two flip chip MADP 000907-14020x PIN diodes [28], D ₁ and D ₂, at two 0.3 mm narrow gaps. A resistor of 5.2 Ω in series with an inductor of 0.45 nH is the equivalent circuit of the diode in the ON state shown in Fig. 3. In the off state, a 7 kΩ resistor and a 0.025 pF capacitor are connected in parallel with the inductor [28]. The biasing circuitry, as seen in Fig. 4, consists of two 100 pF capacitors, C ₁ and C ₂, for dc blocking, and three 250 nH inductors, L_{n 1}–L_{n 3}, for radio frequency (RF) choking. The values of capacitor and inductor are determined with the help of literature survey [Reference Balanis27, Reference Bhattacharjee and Dwari29–Reference Patel and Behera32] and component datasheets [28, 33, 34].

Figure 3.

(a) Equivalent circuit of PIN diode and (b) PIN diode model.

Figure 4.

RF PIN diode used to create a series switch.

RFA design steps and analysis

Fractal generation

Figure 5 displays the fractal structure of the Koch curve. It is a line segment taking parts signified as β. The segment of the part is split into a group of three sub-parts, denoted as β/3. The length of the separated segments of β₁, β₂, and β₃ is L/3. Iteration is the term used to describe this continuous repetition of portion division.

Figure 5.

Koch curve iterations: (a) 0^th, (b) 1^st, and (c) 2^nd order [Reference Pandav, Sadhukhan, Das, Behera and Mohanty20].

The segment is split by a factor of 3 for each iteration of the line L, starting with L ₁ = L/3, continuing with L ₂ = L/9 for the 2nd iteration, L ₃ = L/27 for the 3^rd, and so on until the n ^th iteration. The equations that apply for Koch curve-based structure and iterations to the n ^th order are (1–3). Equation (4) demonstrates that the Koch curve is (4/3)ⁿ for the entire length of the line segment L. Higher-order iterations of the fractal geometry were not feasible due to the limitations and fabrication difficulties. Thus, the 2^nd iteration has been considered the most appropriate iteration of the design in this research endeavor.

(1)\begin{equation}{\text{ }}{\beta _1} = L/3 \to {L_1} = 4\beta \end{equation}

(2)\begin{equation}{\text{ }}{\beta _2} = L/3 \to {L_2} = 16\beta \end{equation}

(3)\begin{equation}{\text{ }}{\beta _n} = L/3 \to {L_n} = {4^n}{\beta _n}\end{equation}

(4)\begin{equation}L = {\left( {4/3} \right)^n}\end{equation}

Larger electrical lengths of radiating patches can be carried out inside a specified region by applying fractal curves along the borders of the regular patch. The Koch curve has drawn a lot of interest within the fractal curve categories because of its simple generation procedure. Here, the asymmetric Koch fractal curves are used to produce two orthogonal modes of similar amplitude for CP radiation along the patch boundaries [Reference Reddy and Sarma15] and [Reference Pandav, Sadhukhan, Das, Behera and Mohanty20]. Figure 2 displays the antenna prototype top and side different points of view. The center of the hexagonal ring patch is filled with a smaller-dimensional Koch fractal element.

Figure 2.

Fabricated prototype of reconfigurable fractal antenna.

Effect of all of the composed iterations

The operating band is directly impacted by fractal geometry that is incorporated into a patch design. The antenna shape is first modelled as a regular hexagonal patch in antenna 1, and then the hexagonal-structured slot is etched from the patch’s middle portion in antenna 2. Furthermore, a second iteration of the Koch curve-based fractal is carried out along the patch element’s inner portion in the antenna construction. Figure 6 shows the 0^th, 1^st, and 2^nd order of iterations of the Koch curve fractal structure. From Figs. 7 and 8, it is observed that, as the number of iterations increases, the depth of operating frequency/axial ratio (AR) band of the radiator increases with a broad frequency range covering 3–7.91/3.33–7.14 GHz but realized gain of the radiator is slightly affected.

Figure 6.

Different stages of reconfigurable fractal antenna.

Figure 7.

S ₁₁ plot for different stages of the antenna.

Figure 8.

(a) Axial ratio and (b) realized gain plot for different stages of the antenna.

Effect of the added extensions

The stages of the evolution of the proposed antenna are depicted in Fig. 9. Their matching performances are displayed in Fig. 10. Two orthogonal modes are formed by the gaps generated by the diodes’ position, which perturbs the electric field [Reference Reddy and Sarma15]. These modes provide CP over 3.33–7.14 GHz. But |S₁₁| is higher than −10 dB. The ground plane represented by Ant. 2 has a slit cut in it with a length of L _f to enhance impedance matching. In Ant. 3, the CP bandwidth is widened by a triangular extension in the ground plane on the right. Consequently, CP bandwidth is significantly enhanced. The dip of AR is shifted and a new dip is created at 6.2 GHz by a second triangular extension on the left. Hence, the CP bandwidth is increased to 3.33–7.14 GHz. The realized gain in the overall band of the antenna increases after cutting the slit and adding left/right triangular extensions to the ground plane, as shown in the Figs. 9 and 10. Initially it is about 2 dBi but after implementing various stages in the antenna, it increased to 5.07 dBi in the operational band. In this case, D ₁ is turned off and D ₂ is turned on for all simulations [Reference Chen, Li, Yang, Cao and Zhang8, Reference Li, Tran and Park10, Reference Pandav, Sadhukhan, Das, Behera and Mohanty20–Reference Bhattacharjee and Dwari22, Reference Abdelgwad35–Reference Kumar, Gupta and Gautam37].

Figure 9.

Different stages of antenna.

Figure 10.

(a) Reflection coefficient (S ₁₁), (b) axial ratio, and (c) realized gain plot for different stages of the antenna.

It is obvious by the above that the ring’s rotating current components are what primarily produce the CP. When D ₁ is ON and D ₂ is OFF, it produces LHCP in the +z direction shown in Fig. 11. RHCP is obtained when the current rotates in the opposite direction shown in Fig. 12 and the diode conditions are inverted, that is, D ₁ is ON and D ₂ is OFF. In a similar manner, two CP modes combine to produce LP when both diodes are ON.

Figure 11.

Surface current distribution of proposed hybrid reconfigurable fractal antenna for Case I (a) at 4.4 GHz (b) at 6.2 GHz.

Figure 12.

Surface current distribution of proposed hybrid reconfigurable fractal antenna Case II (a) at 4.4 GHz (b) at 6.2 GHz.

Effect of ground length L _e

As seen in Fig. 1(c), the parameter L _e indicates the length of the partial ground. Figure 13 illustrates how various L _e values affect the outcome. The length L _e is reduced, the IBW at S ₁₁ less than −10 dB with wideband is achieved. When the length, L _e, decreases, the reflection coefficient at higher resonant frequencies is decreased.

Figure 13.

Reflection coefficient (S ₁₁) plot for different cases of the antenna.

Effect of rectangular slit length L _f in the ground

The effect of rectangular slot length (L _f) of the antenna on the reflection coefficient (S ₁₁) is studied in order to obtain better performance. In Fig. 14, several values of L _f are shown and compared with other values of rectangular slit placed slightly behind the feed on the ground plane when both the switches D ₁ and D ₂ are in state 2 (OFF, ON). An impedance match is created by the addition of a rectangular slit, and coupling between the patch and ground plane occurs. The length (L _f) of the slot is expanded by 1 mm while the slot width (W _f) remains same. The S ₁₁ below −20 dB at 3.5 and 7 GHz rises gradually as L _f increases, as seen in Fig. 14. According to the simulated results, L _f = 4.5 mm has a better return loss than other L _f values because of better impedance matching.

Figure 14.

Reflection coefficient (S ₁₁) plot for different cases of the antenna.

Proposed antenna performance and discussion

The measurement setup is shown in Fig. 15. Figure 16 displays the measured and simulated reflection coefficient (|S ₁₁| < −10 dB) for each of the three polarization states. The S ₁₁ for the CP mode is consistently less than −10 dB between 3 and 7.91 GHz. 3.54–6.75 GHz is the comparable value for the LP mode. A rectangular horn antenna with excellent linear polarization purity is used as the source antenna in the experimental setup depicted in Fig. 15 in order to measure the phase of the two field components with respect to the signal generator. Because rectangular horn antennas produce less cross-polarization over wide bandwidths than circular polarized feeds, a significant source of measurement error is consequently eliminated. The antenna far-field radiation pattern in each cut (Ф) is generated by plotting the parameters ϴ (deg), P _RHCP (dB), and P _LHCP (dB). A practical antenna normally generates a desired reference polarization in addition to an undesirable cross-polar component, which is polarized in the opposite hand. For example, in Fig. 6, RHCP is the reference polarization, and LHCP is the cross-polar component. In the main beam, | P _RHCP (dB)−P _LHCP (dB) | gives the cross-polar level at a given azimuth angle ϴ. For a perfectly circularly polarized pattern, this level is −∞ dB (AR = 0 dB), and for a LP field, where the two CP signals are of identical magnitude, this level is 0 dB (AR = ∞ dB). The reference and cross-polar spatial patterns can also be retrieved from an AR plot, which is generated using a LP spinning source antenna in the experimental setup. To obtain this data, the antenna under test must be moved in azimuth (ϴ) and the source horn must be constantly rotated about its axis (Ф). The ripples in the radiation pattern are a consequence of the beam ellipticity, which occurs when a finite cross-polar component exists. The AR is defined by the depth of the nulls, and it is related to the cross-polarization provided by the formula (5) [Reference Balanis27, Reference Toh, Cahill and Fusco30–Reference Patel and Behera32].

(5)\begin{equation}{\text{AR}} = 20{\log _{10}}\left( {\frac{{1 + e}}{{1 - e}}} \right)\end{equation}

Figure 15.

Measurement setup of the proposed RFA antenna.

Figure 16.

S ₁₁ plot for different cases of the antenna.

where $e = {10^{ - {P_{dB}}/20}}$and ${P_{dB}}{\text{ }}$is the cross-polar power.

The simulated and measured AR for the LHCP and RHCP modes, respectively, is below 3 dB over 3.6–6.9 GHz (62.86%) and 3.33–7.14 GHz (72.78%), as illustrated in Fig. 17. A minor discrepancy between the simulated and measured observations is noticed due to differences in the diode characteristics that are utilized for polarization reconfigurability. Figure 18 displays the measured broadside gain variations. The radiation efficiency of the antenna is analyzed in simulation and measurement mode as shown in Fig. 19. Proposed design achieves radiation efficiencies 94.8/95.9/95.6 and 94.9/97/94.2 (in percentage) during simulation and experimental process in LP/LHCP/RHCP polarization state, respectively. Figures 20 and 21 display the normalized far-field radiation patterns at 4.6 GHz for the CP states, respectively. The sense of polarization is considered in +z direction. The −z direction shows the opposite of it. The difference between the co-polarization and cross-polarization levels is consistently 18.5 dB. Figure 22 displays the radiation patterns for LP. Table 1 provides a summary of the polarization states and corresponding diode states.

Figure 17.

Axial ratio plot for different cases of the antenna.

Figure 18.

Realized gain plot for different cases of the antenna.

Figure 19.

Efficiency plot for different cases of the antenna.

Figure 20.

Radiation patterns of the antenna when D ₁ is ON and D ₂ is OFF (LHCP) (a) at 4.2 GHz and (b) at 5.2 GHz.

Figure 21.

Radiation patterns of the antenna when D ₁ is OFF and D ₂ is ON (RHCP) (a) at 4.2 GHz and (b) at 5.2 GHz.

Figure 22.

Radiation patterns at (a) 4.4, (b) 5.2, (c) 5.8, and (d) 6 GHz.

Table 1.

Operation modes of the proposed wideband polarization reconfigurable antenna (PRA)

The proposed antenna and a few recently published wideband RFAs are compared in Table 2. The switching element of the antenna shown in references [Reference Saraswat and Harish7, Reference Li, Tran and Park10], and [Reference Bhattacharjee and Dwari22–Reference Das, Das, Kumar, Kumari, Mandal, Mukhi and Pradhan24] is positioned inside the radiating element or signal path and requires a sophisticated biasing network in order to switch. With a large total CP bandwidth, the proposed antenna has a good profile. The dual-band antenna mentioned in reference [Reference Huang, Shirazi and Gong23] has performance that is comparable to that of the proposed design, but it is larger and has a switching element inside the RF signal line. Although the narrow band radiator in reference [Reference Das, Das, Kumar, Kumari, Mandal, Mukhi and Pradhan24] has a low gain, it is larger in size and has a lower overlapping CP bandwidth than the proposed antenna. In Table 2, few selected designs are having low ARBW. Larger sizes are yet another significant disadvantage of these radiators. On the other hand, the proposed design operates in different polarization states, wide IBW and compact in size. This illustrates the benefits of the RFA above previous comparable works. The majority of antennas intended for reconfigurability will have complex design and multiple no. of switching elements. However, the designed antenna uses the fewest possible switching components. The primary goal of this research is polarization reconfiguration of antenna for wireless applications. The significant gain at the operating frequency bands is provided by the antenna. The antenna covers a wide range of application bands, and polarization reconfiguration are offered by applying two PIN diodes. The designed structure is distinguished because it is compact in dimensions, has a good substrate material (Rogers RT Duroid), and has a peak realized gain of 5 dBi across the whole realized gain compared to other structures. Table 2 shows that the proposed fractal radiator has a high gain, small dimension, and better polarization reconfiguration performance, making it suitable for wireless communication services.