I. INTRODUCTION
An explosive growth of wireless radio communication systems is currently observed in the microwave band. In [Reference Xuzhe, Hua, Huaiwu and Zhiyong1], a dual-layer micro-strip antenna working at 1.268 GHz was designed and analyzed. The bandwidth of the micro-strip antenna was 35 MHz, and the gain was 6 dB. In [Reference Guanlong2], a novel broadband microstrip patch antenna was proposed and studied. The antenna achieves broadband by adopting a bilayer patch for coupling; a capacitive metal-wafer is introduced at the top of the feeding probe to counteract the inductance caused by the probe. In addition, Minkowski fractal-patch structure was applied in antenna design to realize its miniaturization. The reflection coefficient is lower than −10 dB at 3.33–4.0 GHz and the impedance bandwidth is 18.3%. Moreover, antenna gain is stable and higher than 6.5 dBi. In [Reference Jeon5], wideband dual-polarized microstrip antennas were studied. The experimental results showed that an impedance bandwidth (SWR ≤ 2) of 33.24% and a peak gain of 8.44 dBi (at 2.11 GHz) were obtained. They were designed originally for the triple service band. In [Reference Zhang, Zhang, Lin and Zhao6], a stacked microstrip antenna with aperture coupling feeding as the individual radiator of the wide band antenna array was studied. With adjusted parameters, it exhibits a broad impedance bandwidth (VSWR ≤ 2) of about 27% (8.8–11.6 GHz).
In [Reference Bhatnagar7], new electromagnetic coupling fed low profile broadband high gain E-shaped microstrip antennas (MSA) were proposed for high speed wireless networks in IEEE 802.11a and standards. The proposed antenna uses an E-shaped microstrip patch covered by a radome and fed by an electromagnetically coupled strip. To validate this concept, a single antenna element and a sub-array were designed, built, and measured. The measured results indicate that the element and the sub-array cover the band from 4.8 to 6.0 GHz (return loss <−10 dB) and produce a gain of 8 and 11 dBi, respectively. In [Reference Yikai, Shiwen and Zaiping8], a simple bandwidth enhancement method for low profile E-shaped patch antennas was presented. By introducing a distributed LC circuit to the E-shaped patch antenna, a new resonant frequency close to that of the E-shaped patch was obtained; measured results show that the designed low profile antenna achieved an impedance bandwidth over 9% for VSWR < 2. In paper [Reference Jay nthy, Sugadev, Ismaeel and Jegan12], a multiple slot microstrip patch antenna for wireless communication was presented, inverted microstrip patch antenna, direct probe feed technique, and the novel multiple shaped patch is used in antenna design. The multi-slotted microstrip patch antenna achieved a fractional bandwidth of 27.89% (1.82–2.41 GHz) at 10 dB return loss. Microstrip antennas were found to be favorable because they were (1) low profile and (2) inexpensive to manufacture and compatible with monolithic microwave integrated circuit designs (MMIC). However, they also had drawbacks such as (a) low efficiency and (b) narrow impedance bandwidth. The input impedance of MSDA depends on its geometrical shape, dimension, and feed type. Therefore, antenna input impedance is an important design parameter that controls the radiated power and the impedance bandwidth. In most applications, bandwidth limitations occur due to an impedance mismatch. In this respect, a microstrip antenna has narrow bandwidth because of heavy reactance. As microstrip antennas have found a wide variety of application areas, a number of techniques have evolved to improve their limited bandwidth. A straightforward application to improve the bandwidth is by increasing the thickness of the substrate supporting the microstrip patch. However, limitations exist with the ability to effectively feed the patch on a thick substrate and the radiation efficiency can degrade with increasing substrate thickness. Techniques for overcoming these band limiting problems can be achieved by using parasitic term elements, external matching, and separating the feed and antenna. The proximity coupling method uses an impedance matching stub connected to the feed line achieving 13% bandwidth. A broadband rectangular patch antenna with a pair of wide slits of near resonance frequency improves to 24% bandwidth. Bandwidths of 10–12% can be obtained using the passive coplanar matching network. However, the gain and radiation pattern is distorted. Using an aperture-coupled antenna, bandwidths larger than 20% are achieved.
II. PROCEDURE FOR OPTIMIZATION
In this paper, we present an optimized microstrip patch antenna for achieving large gain over large bandwidth, using change in the position of parasitic patches on the top layer.
A) Optimization by placing two parasitic patches on the top layer as shown in Fig. 1
The proposed model of optimization Section A is shown in Fig. 3 and 3D view of Proposed Model of optimization Section A is shown in Fig. 2. We optimized spacing S between top patches, using two parasitic patches on the left side of the geometry, and verified the result for each spacing and appropriate position. After a number of iterations, appropriate spacing S = 2λ was achieved.
Proposed design of Section II(A).
3D view of Section II(A).
Proposed model of Section II(A).
We optimized the following points and achieved an appropriate value.
a. Slot loaded on parasitic patch.
b. Δ = 3 mil air gap between layers.
c. Whole geometry consists of the layer glass epoxy PCB and air gap (FR-4–air–FR-4).
d. Total height of geometry is 121 mil (from a ground plan to top layer).
e. Spacing between parasitic patches is S 1 = 62 mil.
f. Spacing between active and parasitic patches is S = 23.62 mil (0.02λ).
g. Middle layer consists of patch dimension L × W = 411 × 446 mil2.
h. Top layer consists of multilayer parasitic patches of the same dimension L 1 × W 1 = 100 × 150 mil2.
i. Optimization by changing the position of top parasitic patches.
The top layer dimension of optimization section A is shown in Fig. 4.
Front view of optimization design for S = 2λ.
From the first optimization section, we achieved optimal directional gains of 6.4–8 dBi over a 5.9 GHz bandwidth slot (8.5–14.5 GHz). The first optimization section includes U-slots on the top patches for impedance matching over a large bandwidth, the gap spacing between top patches is S = 2λ, the total height of the proposed antenna is 121 mil by using a multilayer concept and the air gap between the top and bottom layers substituted the 411 × 446 mil2 patch on the middle layer of geometry and the feed probe connected to the middle layer. The current density on the top parasitic patches depends on the geometry of top patches 100 × 150 mil2, cutting slot length, width of the slot, and the gap between parasitic patches. We have observed strong coupling between top patches and achieved optimal gains. The results of this optimization are shown in Figs 5–7.
Return losses versus frequency.
VSWR versus frequency.
Directivity versus frequency.
We have achieved VSWR ≤ 2 over a 5.9 GHz impedance bandwidth (8.5–14.5 GHz), and −10 dB impedance bandwidth over a 5.9 GHz bandwidth slot. However, in terms of directional gain and bandwidth, we require further modifications in the result so that the position of parasitic patches is changed on the top layer and the observed results achieve 34% impedance bandwidth of the X-band and 13.4% impedance bandwidth of the Ku-band as shown in Fig. 5.
B) Effect of two parasitic patches: one at the left side bottom corner and another at the right side top corner
For enhancing the flat gain curve over large bandwidth, we require further modifications in the previous design so that we can arrange one parasitic patch at the right side top corner and the second patch at the left side bottom corner as shown in Fig. 8. 3D View of Section B is shown in Fig. 9. From this arrangement, the left side patch provides impedance bandwidth for the lower edge of the frequency and the topside patch provides impedance bandwidth for the higher edge of the frequency due to which we have achieved a 5.7 GHz broad bandwidth slot. From optimization section B reduced mismatching loss, ohmic loss and circular polarization loss. Losses are reduced due to removing most of copper and placing two U-Slots on the cross arrangement on top layer. Overall, we achieved 6.2–8.8 dBi over a broad bandwidth 5.7 GHz (8.5–14.5 GHz). All results shown in Figs 10–12 achieved VSWR ≤ 2 over a 5.7 GHz bandwidth slot, −10 dB impedance bandwidth up to 5.7 GHz and 38% impedance bandwidth of the X-band and 12.77% impedance bandwidth of the Ku-band as shown in Fig. 10.
Modified design for Section II(B).
3D view of Section II(B).
Return losses versus frequency for Section II(B).
VSWR versus frequency for Section II(B).
Directivity versus frequency.
C) Effect of two parasitic patches: one at the left side of the top corner and another at the right side of the bottom corner
In section C, we have arranged one parasitic at the right side bottom corner and second at the left side top corner shown in Fig. 13 and 3D View of Section C is shown in Fig. 14. From this optimization, we achieved a 5.8 GHz broad bandwidth slot, the left side parasitic patch provided impedance bandwidth for the lower edge of the frequency and the topside parasitic patch provided impedance bandwidth for the higher edge of the frequency. The position of patches reduced surface loss, cross polar loss; achieved more flat directional gains 6–8 dBi over a broad bandwidth of 5.8 GHz (8.5–14.5 GHz) shown in Fig. 17, achieved VSWR ≤ 2 over a 5.8 GHz bandwidth slot, −10 dB impedance bandwidth up to a 5.8 GHz slot of bandwidth. All the results are shown in Figs 15–17; we achieved 38% impedance bandwidth of the X-band and 13.36% impedance bandwidth of the Ku-band as shown in Fig. 15.
Modified design for Section II(C).
3D view of Section II(C).
Return loss versus frequency for Section II(C).
VSWR versus frequency for Section II(C).
Directivity versus frequency.
D) Effect of four parasitic patches: two at the top corner and another two at the bottom corner
We arranged two parasitic patches at the right side corners and two other parasitic patches at the left side corners as shown in Figs 18 and 19. From this arrangement, we achieved a 6 GHz broad bandwidth. The position of the patches reduces surface loss, cross polar loss, and ohmic loss. We achieved a very high flat directional gain of 6.2–8.8 dBi over a broad bandwidth of 6 GHz (8.5–14.5 GHz) as shown in Fig. 20. All the results are shown in Figs 20–22; we achieved 38% impedance bandwidth of the X-band and 14.8% impedance bandwidth of the Ku-band as shown in Fig. 18.
Modified design for Section II(D).
3D view of Section II(D).
Directivity versus frequency.
Return loss versus frequency for Section II(C).
VSWR versus frequency for Section II(C).
We have further discussed the method of fabrication. For the fabrication, FR-4 and air are used; firstly by designing the layout in AutoCAD software, etching is completed using ferric-chloride acid, and for feeding the SMA Connector is used. The fabricated proposed antenna is shown in Fig. 23.
(a) Bottom view of fabricated proposed antenna. (b) Top view of fabricated proposed antenna.
III. RESULTS AND DISCUSSION
We have focused on optimal directional gain and broad bandwidth. For achieving these outcomes, we have used the genetic algorithm on an IE3DTM Simulator. The impedance frequency bandwidth of a microstrip antenna depends primarily on the thickness, the dielectric permittivity of the substrate, position, and dimension of the parasitic patch on the top layer. We have analyzed four cases. From the first case (Section II(A)) we achieved 5.9 GHz impedance bandwidth, from the second case (Section II(B)) we achieved 5.7 GHz impedance bandwidth, from the third case (Section II(C)) we achieved 5.8 GHz impedance bandwidth, and from the fourth case (Section II(D)) we achieved 6 GHz impedance bandwidth. All geometry consists of a parasitic array concept. We have studied stack geometry with a number of active and parasitic patches for enhancing the maximum directional gain and reduction in surface wave, cross polarization loss. We have achieved good impedance matching due to all aspects of modeling and achieved maximum directional gains of 6.2–8.8 dBi from Section II(D). All optimization section results are shown in Table 1.
VSWR versus frequency.
IV. CONCLUSION
From the optimization of four sections (Sections II(A)–II(D) in the IE3D™ Simulator using the genetic algorithm, we obtained appropriate geometry. The Geometry Section II(D) reported a 6 GHz S11 < −10 dB impedance bandwidth slot; we achieved 38% impedance bandwidth of the X-band and 14.8% impedance bandwidth of the Ku-band. The optimization section has two parasitic patches, one at the top corner of the left side and one at the bottom corner of the right side, and has reported 5.9 GHz S11 <−10 db impedance bandwidth slots. Overall we have been concluded that the geometry has two parasitic patches in the left side corner and two parasitic patches in the right side corner is optimum. The optimization section of four parasitic patches (Section II(D)) provided a high flat gain response of 7.5 dB, over a 6 GHz bandwidth slot (8.5–14.5 GHz). We performed optimization in four steps, by changing the position of parasitic patches on the top layer, and analyzed results on the IE3D™ Simulator. The appropriate four parasitic patches on top layer antenna design are useful for X–Ku band application, used as a Satellite and Radar antenna at the X–Ku band. The hardware validation and simulation results are matched.
ACKNOWLEDGEMENT
This work was supported by Lambda Microwave Pvt. Ltd for hardware validation.
Dr Anubhuti Khare (BE, MTECH, Ph.D.) is a Professor in the Electronics and Communication department, UIT RGPV Bhopal. She has published 50 papers in international journals. She has guided many theses for PG students and Ph.D. scholars. She has 20 years of teaching experience at the PG and UG level.
Rajesh Nema (BE, MTECH, pursuing a Ph.D.) is a Ph.D. scholar at UIT RGPV Bhopal. He has published 10 papers in international journals.
