Reflection loss

Reflection loss is characterized as the measure of power loss caused by the reflection of power at a line interruption. It is communicated in terms of decibels (dB) and the return loss is conveyed in equation (1).

(1)$$R_L( {\rm dB}) = {-}20\log 10( \tau ) , \;$$

where τ is the reflection loss of the proposed antenna.

The return loss of the proposed elliptical-shaped patch antenna and the other two conditions sustained with MEMS switch ON and OFF state is shown in Fig. 4. The reconfigurability can be observed in the figure at different conditions of the switch. The proposed antenna exhibits maximum reflection loss at the antenna with switch in OFF state. The performance of the fabricated prototype is measured by a vector network analyzer in an anechoic chamber presented in Fig. 5. The comparison analysis of the proposed antenna's simulated and measured results is shown in Fig. 6.

Fig. 4.

Reflection loss of the proposed antenna with different states of the RF-MEMS switch.

Fig. 5.

(a) Fabricated antenna with vector network analyzer in an anechoic chamber. (b) Fabricated prototype of the proposed antenna design.

Fig. 6.

Simulated and measured results of the proposed antenna design.

Impedance matching

Impedance coordination is presented by employing a coordinating system with lumped components (LandC). Different feasible improvements are accessible for coordinating system design. They are mostly influenced by design complication, its execution, and flexibility. The ideal plan for matching network utilizing lumped components can be achieved by Smith chart. The ideal impedance of the antenna (Z _L) is ought to be considered. Eventually, the ideal impedance for the switch can be validated as (Z ₀). In order to meet the required assessment, the device geometry is to be made properly. To attain the proper matching, the antenna and the switch need to satisfy the condition (Z ₀ = Z _L). The lumped elements involve in matching the input impedance. Only when the proper matching is done, it can be verified with the Smith chart. Though many solutions are available, only Smith chart is selected among them. To verify the proper matching, the input impedance should meet the focal point of the Smith chart. If the input impedance is away from it, the condition mentioned above will not be satisfied. The impedance ought to be considered as 50 Ω.

From Figs 7–10, the characteristic impedance of the antenna, antenna with switch ON state, antenna with switch OFF state, and finally the proposed switch is observed. The investigation on impedance with and without RF-MEMS switch on elliptical patch antenna is studied. The switch may be in open or closed condition depending on the scattering parameters.

Fig. 7.

Impedance matching of the antenna in Smith chart showing an input impedance of 50 ohms.

Fig. 8.

Elliptical-shaped microstrip patch antenna with the proposed switch in ON-state impedance-matching network in Smith chart.

Fig. 9.

Elliptical-shaped microstrip patch antenna with the proposed switch in OFF-state impedance-matching network in Smith chart.

Fig. 10.

Impedance-matching network of the proposed switch in Smith chart.

The impedance of the elliptical patch antenna and the embedded switch is observed for two (ON and OFF) states. The characteristic impedance of the switch matches with the proposed antenna impedance.

Radiation patterns

The implementation of the elliptical-shaped antenna is studied for various far field conditions. The following figures represents the radiation patterns of the proposed antenna with respect to E plane an H plane. Here, three distinctive far field conditions are used, those are φ = 0, φ = 90, and the outlines for these conditions are observed in Figs 11(a-c), respectively. Radiation of the proposed antenna at the frequency of 10.46GHz, the proposed antenna with MEMS ON state at the frequency of 10.57GHz, the proposed antenna at the frequency of 10. 53GHz.The radiation patterns of all the three conditions in E plane are observed as omni directional.

Fig. 11.

(a) Patterns of the proposed elliptical-shaped antenna. (b)Patterns of the proposed elliptical-shaped antenna with switch in ON state. (c) Patterns of the proposed elliptical-shaped antenna with switch in OFF state.

The normalized radiation patterns of the proposed elliptical patch antenna are shown in Fig. 11(a). The simulated patterns of the antenna are matched well with the measured results at 10.46 and 10.5 GHz, respectively. Due to measurement errors, there is a slight difference between the simulated and measured results.

Figures 12 and 13 show the current distributions of the proposed elliptical microstrip patch antenna with and without switch. The losses are minimized due to elliptical shape.

Fig. 12.

Surface current distributions of the proposed antenna without switch at 10.46 GHz.

Fig. 13.

Surface current distributions of the proposed antenna with switch (OFF-state) at 10.57 GHz.

Figure 14 shows the simulated and measured radiation efficiency of the elliptical microstrip patch antenna. The deviation from simulated and measured results is due to issues raised during fabrication. Radiation efficiency is presented to validate the antenna performance.

Fig. 14.

Radiation efficiency of the elliptical patch antenna.

The simulated and measured gains of the proposed antenna are 6.13 and 5.9 dB, respectively, as shown in Fig. 15. For a better understanding of the proposed model antenna and higher gain obtained when compared to the other antenna models, see Table 3.

Fig. 15.

Gain versus frequency of the elliptical patch antenna.

MEMS switch analysis

In the process of design, the actuating voltage applied on the switch is unique which needs to be low and considered to be a challenging parameter in RF-MEMS; usually when the switches are activated, the metal beam is bent down and restores its elasticity to the original position. The beam in the upstate is subjected to duty electrostatic force and thus pull-in voltage expression can be given as:

(2)$$V = \sqrt {\displaystyle{{2k} \over {\varepsilon _0A}}g^2} ( g_0-g) , \;$$

where g is the maximum air gap maintained between the plates, k is the spring constant, and A is the overlapping area of beam and lower electrode when actuated; here, when

(3)$$\displaystyle{{\partial v} \over {\partial g}} = 0\Rightarrow g = \displaystyle{{2g_0} \over 3}, \;$$

representing the maximum actuation voltage when the air gap between the upper and lower electrode reaches 1/3rd of its height, then the expression can be evaluated as

(4)$$V_P = \sqrt {\displaystyle{{8k} \over {27\varepsilon _0A}}} g_0^3 , \;$$

where A is the area of overlap.

Actuation voltage is defined as the voltage that is given to the upper beam and lower electrode.

RF performance of the device is evaluated with lumped elements. The typical values of the lumped components are important to notify. The ON and OFF states of the switch capacitance are the important ones which are given in the equivalent circuit. The ON-state capacitance is evaluated with the expression.

In the parallel plate switches, charges are implemented through charge carrier. Based on the overlapping area, parallel plate capacitance relies between the two capacitive plates. The total beam area is directly proportional to the total switch capacitance. Here, the term C_pp represents the ideal capacitance (parallel plate). It is important to support the minimal capacitance to attain the required frequency for the switch. The C_ff simply relies upon the holes on the switch beam. Here, in the proposed shunt switch, the electric field moves across the switch beam and modifies the overall capacitance of the switch beam. However, the C_ff can be observed across the upper surface of the plate and between the sides of the plates.

(5)$$C_{ON} = C_{\,pp} + C_{\,ff} = \displaystyle{{\varepsilon _0 + A} \over {g_0 + ( {t_d/\varepsilon_r} ) }}, \;$$

where C _pp is the parallel plate capacitance and C _ff is the fringing field capacitance.

When the switch is in ON state, the on capacitance must be low in order to keep the insertion loss minimum. The OFF-state capacitance is typical to calculate because when the membrane is in OFF state, it is not fully flat over the oxide layer. The switch OFF-state capacitance can be expressed as

(6)$$C_{OFF} = \displaystyle{{\varepsilon _0\varepsilon _rA} \over {t_d}}.$$

The ON-state and OFF-state capacitance of the switch with aluminum nitride as a dielectric is 1.83 and 2.91 pF. The ON-state capacitance should be lower than the OFF-state capacitance.

The deformation of beam with different beam thickness at 0.8, 1, and 1.2 μm is considered to evaluate the switch performance. To validate the switch three different beam thickness like 0.8, 1 and 1.2 µm are considered.and analysed) instead of (To validate the switch and for the material saving three analysis are ought to be considered for the switch after 1µm if the thickness is increased the beam cannot meet the required demand. The time required for a switch to toggle from ON state to OFF state can be expressed as

(7)$$t_s = \displaystyle{{3.67V_p} \over {V_s\omega _0}}, \;$$

(8)$$\omega _0 = 2\pi f_o, \;$$

(9)$$f_o = \displaystyle{1 \over {2\pi \sqrt {LC} }}.$$

There are switching times concerned with switch. The first referred to pull-in time or the settling time, defined as the time taken to move the flexible beam to the bottom ground electrode through the voltage applied referred as pull-in voltage. The second referred to the pull-up time or the recovery time, the time taken by the beam to release from the pull-in position to its switched position by removing the voltage applied to the beam referred as pull-up voltage. Here the settling time and the recovery time of the switch are presented in Fig. 16. Due to the force applied on the switch, it undergoes three different stresses at overall, center and edge of the beam.

Fig. 16.

Switching time and recovery time of the proposed switch with different states (ON and OFF): (a) all over the beam, (b) at the center of the beam, and (c) at the edges of the beam.

In this section, the analysis is presented with switch on the proposed antenna. Switch performance is evaluated on the isolation, low return, and insertion loss. Therefore, by adjusting the geometrical parameters of the switch, the RF performance can be optimized. It is concerned with the return loss of −78 dB at 8 GHz and the insertion loss of −0.07 dB at 8 GHz frequency. By changing the dielectric height, the capacitance between the electrodes is changed, leading resonant frequency change. So, the dielectric layer thickness must be chosen between 0.1 and 0.5 to attain the required frequency range. The isolation performance of the proposed switch can be observed in the following figures. The isolation of −77 dB at the frequency of 8 GHz is obtained. From Figs 17 and 18, the performance can be observed. The actuation voltage calculated for the proposed switch is 7.9 V to actuate the top electrode to move from ON state to OFF state.

Fig. 17.

Isolation performance of the switch.

Fig. 18.

(a) Return loss and insertion loss of the proposed switch. (b) Actuation voltage and displacement curve.

The performance of the RF-MEMS switch related to the low return loss, insertion loss, and the improved isolation is −78, 0.07, and −77 dB at 8 GHz. The radiation patterns of the proposed elliptical antenna integrated with the switch are observed. The proposed SRR elliptical patch antenna is compared with the previous works in Table 3.

Table 3.

Comparison of the proposed SRR elliptical patch antenna with literature work

N/A, not available.

The proposed antenna is suitable for satellite applications and the optimization is carried out with HFSS. So, we consider only one resonant frequency band. The purpose of generating a single resonant band is reducing interference because the satellite communication antenna may have low losses. However, the proposed shape can be generated by a dual band with different optimization techniques.