A circularly polarized slot antenna for high gain applications

A circularly polarized slot antenna for high gain applications

Accepted Manuscript Title: A Circularly Polarized Slot Antenna for High Gain Applications Author: R.V.S. Ram Krishna Raj Kumar Nagendra Kushwaha PII: ...

2MB Sizes 0 Downloads 4 Views

Accepted Manuscript Title: A Circularly Polarized Slot Antenna for High Gain Applications Author: R.V.S. Ram Krishna Raj Kumar Nagendra Kushwaha PII: DOI: Reference:

S1434-8411(14)00166-6 http://dx.doi.org/doi:10.1016/j.aeue.2014.05.018 AEUE 51228

To appear in: Received date: Revised date: Accepted date:

16-10-2013 30-5-2014 31-5-2014

Please cite this article as: Krishna RVSR, Kumar R, Kushwaha N, A Circularly Polarized Slot Antenna for High Gain Applications, AEUE - International Journal of Electronics and Communications (2014), http://dx.doi.org/10.1016/j.aeue.2014.05.018 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A Circularly Polarized Slot Antenna for High Gain Applications

cr

ip t

RVS Ram Krishna, 1Raj Kumar and Nagendra Kushwaha Research Scholar,Department of Electronics Engg.,DIAT(Deemed University), Pune-411025,India 1 ARDE, Pashan, Pune – 411 021,India

d

M

an

us

Abstract – A coplanar waveguide fed slot antenna for wideband circular polarization is designed and experimentally validated. The rectangular slot is excited using a stepped feed line terminated on a circular disc shaped tuning stub. To obtain circular polarization, inverted L-shaped strips are attached to the ground plane at the opposite corners while a rectangular slit is cut in the circular disc. The combined bandwidth (3-dB axial ratio and 10 dB impedance matching) achieved is 48% (4.35 GHz to 7.1 GHz) under simulation and 40% (4.75 GHz to 7.1 GHz) in measurement. The gain of the antenna is next enhanced by the application of a double layered square loop frequency selective surface. The frequency selective surface is used as a reflector placed beneath the antenna at an optimum distance. An improvement of about 4 dB is seen in the measured peak gain over most of the operating band. Experimental results are presented to characterize the antenna and the frequency selective surface and they are found to be in good agreement with the simulated results.

Ac ce p

te

Keywords ; Slot Antenna, UWB Antenna, Circular Polarization, High Gain, Frequency Selective Surface

I. Introduction

There is an ever increasing demand to improve the data transfer rate and enhance the transmission quality of UWB wireless communication systems. In this respect, printed antennas with dual polarization and circular polarization characteristics can be used. A circularly polarized antenna allows for flexible orientation of the transceiver and also helps to combat multi-path fading effects in a diverse environment [1]. A number of printed antennas displaying circular polarization (CP) properties have been reported in the published literature. In particular, the coplanar waveguide (CPW) fed slot antenna is preferred for its cost efficient uniplanar structure and wider bandwidth. The uniplanar structure also reduces misalignment errors to a large extent [2]. In addition, the CPW feed is characterized by less dispersion, low radiation loss and ease of integration with monolithic microwave integrated circuitry. Some of the slot antennas recently reported for circular polarization are listed in references [3 – 8].

Page 1 of 21

an

us

cr

ip t

Along with a desired polarization profile, another feature of the antenna quite important in several applications is the gain or directionality property. As against the microstrip fed antennas, the CPW fed monopoles or slots suffer from reduced gain due to the absence of a ground backing. The radiation patterns of these antennas tend to be omni-directional implying a wastage of power for applications such as point to point communications or object tracking. Other applications where the extended bandwidth of UWB is used such as microwave imaging, biomedical imaging, non destructive detection and ground penetrating radars also require high gain or uni-directionality. The uni-directional feature also improves the signal to noise ratio. One solution to improve the gain of a CPW feed antenna is to provide a metallic reflector beneath the antenna which can also act as a shield for the adjacent electronic circuitry. However, the metallic shield causes image currents to appear and the out of phase reflections result in deterioration in the impedance matching and distortion in the far-field radiation patterns. Another solution is the application of a frequency selective surface (FSS) just like a reflector. The FSS is also called a high impedance surface (HIS) and when properly placed, can offer in-phase reflection over a wider band and improve the impedance matching. In other words, it can be used to enhance the gain and the bandwidth of an antenna [9-13].

Ac ce p

te

d

M

In the work presented in this paper, a compact slot antenna utilizing CPW feed is designed. The geometry of the slot is a rectangle with perturbations applied for getting circularly polarized behavior. The CP bandwidth realized is 40% centered at 6 GHz while the impedance bandwidth attained is 121% (from 2.5 to 10.2 GHz). To improve the gain of the antenna, a frequency selective surface is designed and implemented. The FSS is a square loop printed on both sides of substrate and placed beneath the antenna to act as a reflector. The double layer is chosen since a multi layer FSS offers wider bandwidth and sharper roll-off [14-15]. The gain improvement with the FSS antenna combination is around 4 dB. It is observed that with this FSS, though there is an improvement in the gain, there is a loss in CP bandwidth. To retain the CP bandwidth, another patch type FSS is designed and simulation results are presented. The software used for simulation is CST Microwave Studio (CST MWS) while measurements are taken using a Rohde and Schwarz Vector Network Analyzer (R&S- ZVA 40). In the following section, the antenna geometry is described followed by a section on simulated and measured results. Thereafter, the FSS design is described and the enhancements in the gain and the radiation patterns are discussed. Finally conclusions are made.

Page 2 of 21

II.Antenna Configuration

Ac ce p

te

d

M

an

us

cr

ip t

The geometry of the proposed antenna and its placement in the co-ordinate system is depicted in Fig. 1(a). A glass epoxy (FR4) substrate of relative permittivity r = 4.4, loss tangent tan  = 0.002 and thickness 1.58 mm is utilized for fabricating the antenna. The ground plane is printed on one side of the substrate and measures W x L mm2. A rectangular slot is etched on the ground plane which has dimensions Ls x Ws mm2. The width of the metallic strip surrounding the slot is ‘p’ mm for the vertical section and ‘q’ mm for the horizontal section. As seen from the figure, the rectangular slot is excited by a CPW whose feed line is two - stepped for better impedance matching. The widths of the two sections of the feed line are indicated by w1 and w2 whereas the length of the section closer to the port is denoted by ‘ls’. The feed is terminated on a circular disc shaped patch protruding into the slot center. The patch has a radius of ‘R’ mm and works like a monopole. For obtaining circular polarization characteristics, inverted L-shaped strips, 1 mm wide are attached to the ground plane at the opposite corners of the rectangular slot. The lengths of the various sections of these strips are indicated by ‘a’, ‘b’, ‘c’ and‘d’. Further, a slant rectangular slit is cut in the circular disc. The slit measured 1 mm in width, 7.8 mm in length and is inclined at 450 with the vertical. All the parameters of the proposed antenna are listed with their values (in mm) in Table I. A photograph of the fabricated antenna is shown in Fig. 1(b).

(a)

(b)

Fig. 1(a) Geometry of the proposed antenna (b) fabricated prototype

Page 3 of 21

Value (in mm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

L W Ls Ws p q R W1 W2 a b c d e g h t

40 40 24.5 30 5.0 3.5 6.0 3.6 2.8 9.0 5.0 5.0 2.5 7.8 0.5 1.2 1.6

an

M

d

Ac ce p

te

cr

Parameter

us

S.No

ip t

Table I Optimized Antenna Dimensions

III. Simulated and Measured Results

The proposed antenna was initially simulated on CST Microwave Studio and then fabricated with the optimized dimensions. The measured and simulated reflection coefficients are compared as shown in Figure 2 and seem to be in good agreement. The impedance bandwidth seen from the simulated reflection coefficient (for S11 < -10 dB) starts from 2.2 GHz and extends well beyond 12 GHz. The measured S11 characteristic follows the simulated S11 however at higher frequencies (> 9 GHz) the measured return loss increases due to the quality of the SMA connector employed and the measured impedance bandwidth is restricted up to 10.2 GHz. From the simulated reflection coefficient (dashed – blue line, Fig. 2), resonances can be noted at 2.75 GHz, 4.0 GHz, 5.4 GHz and 9.6 GHz. The first resonance is controlled by the wide slot of dimensions Ws x Ls; the slot perimeter being approximately equal to one guided wavelength at this frequency. The circular disc at the end of the CPW line acts like a monopole and results in the resonance at 4.0 GHz. The open slit cut in the circular disc causes the resonance at 5.4 GHz while the higher resonances seem to

Page 4 of 21

an

us

cr

ip t

originate in the tank circuits formed by the inverted L-shaped strips attached at the corners of the rectangular slot. A plot of the surface current distribution at these resonances is shown in Fig. 3 which justifies the above observations.

Ac ce p

te

d

M

Fig. 2. Measured and simulated reflection coefficients of the proposed antenna

Fig. 3. Surface current distribution at the resonance frequencies of the slot antenna

Page 5 of 21

Ac ce p

te

d

M

an

us

cr

ip t

The axial ratio of the proposed antenna is measured inside an in-house anechoic chamber using Antenna Measurement System in conjunction with a Vector Network Analyzer (R&S, ZVA-40). The measured and simulated axial ratios along a direction  = 730 and  = -890 in which the best performance (maximum ARBW) is obtained are shown next in Fig. 4. The simulated axial ratio stays less than 3 dB from 4.35 GHz and up to 7.1 GHz indicating an axial ratio bandwidth of 48% centered at 5.72 GHz. The measured axial ratio bandwidth (ARBW) is seen to be slightly less than the simulated ARBW at 40% (4.75 GHz to 7.1 GHz). The wideband CP is achieved by shape optimization of the slot and the tuning stub and the perturbations employed (inverted L-shape strips and the slant rectangular slit on the circular disc) which control the polarization properties of the antenna. The L-shape strips attached to the ground plane create alternate current paths. Currents flowing through the bent sections will be in space quadrature to the normally flowing ground return currents. The time quadrature is created by the inductance of the metallic strips. Thus both the conditions for circular polarization are fulfilled. In particular, simulation studies have shown that the horizontal length of the upper L-shaped strip controls the axial ratio to a large extent and has to be optimized for the best performance. The length and inclination angle of the slant rectangular slit further enhance the CP behavior.

Fig. 4. Measured and simulated axial ratio (along  = 730 and  = -890) of the proposed antenna To assess the spread of the CP characteristics from this optimal direction ( = 730,  = -890), a 2D plot of the axial ratio (versus  and ) at different frequencies within the simulated CP bandwidth is shown in Fig. 5. From the antenna geometry (Fig. 1) it can be inferred that positive values of phi point to the right half of the antenna when viewed from the top and negative values of phi point to

Page 6 of 21

te

d

M

an

us

cr

ip t

the left half of the antenna. From Fig. 5 it can be seen that at lower frequencies within the AR bandwidth, CP spread is better on the left side (  < 00 ) whereas at higher frequencies it is better on the right side of the antenna (  > 00 ). The asymmetry can be attributed to the different lengths and different positions of the L-shape strips attached to the two halves of the slot. In Fig. 6, the variations in the cross polarization (the ratio of LHCP to RHCP) with different directions are shown. The deviations taken are ± 100 (in both  and ) from the chosen direction. Although, the cross polarization increases with a change in the direction (as expected), it remains low (<-10 dB) over the CP bandwidth for the tested deviations (±100).

Ac ce p

Fig. 5. 2D plot of the axial ratio (versus  and ) at different frequencies (4.6 GHz, 5.0 GHz, 5.4 GHz, 6.0 GHz, 6.4 GHz and 7.0 GHz) within the CP bandwidth.

(a) (b) Fig. 6. Simulated cross polarization (LHCP/RHCP) for different directions; (a) ± 100 deviation in  from optimal direction (b) ± 100 deviation in  from optimal direction (=730, =-890)

Page 7 of 21

IV. FSS Design for Gain Enhancement

cr

ip t

A 6 x 6 square loop FSS is designed to improve the gain of the proposed antenna. The FSS is formed by printing metallic loops on both sides of FR4 substrate of relative permittivity r = 4.4 and thickness 1.6 mm. A schematic of the FSS is shown in Fig. 7. The simulated and measured transmittance for the FSS is shown in Fig. 8. From Fig. 7, the following dimensions of the FSS screen can be easily inferred.

us

Table II. Dimensions of the FSS Denotation

Width of the metallic loop Dimension of the Unit Cell Periodicity Gap

w d p g

Value (in mm) 3.5 14.0 14.5 0.5

Ac ce p

te

d

M

an

Parameter

Fig. 7(a) Schematic of the proposed FSS (b) dimensions of the unit cell and periodicity

Page 8 of 21

ip t cr us an

M

Fig. 8. Measured and simulated transmittance of the square loop FSS. Also shown is the simulated reflection phase.

Ac ce p

te

d

The loop type FSS combines the characteristics of the patch type FSS and the slot type FSS and is chosen for its broad band operation. By using a two layered structure with the loop FSS, ultra wide bandwidth can be realized. As a comparison, the transmittance with a single layer square loop FSS is shown against the transmittance with a double layer square loop FSS in Fig. 9(a). It is easily seen from the figure that a wider bandwidth of 7.2 GHz (3.9 GHz –11.10 GHz) is realized in case of the latter.

Fig. 9(a). Simulated transmittances for the square loop FSS (single layer and double layer).

Page 9 of 21

ip t

cr

Fig.9(b). Equivalent circuit for the double layer, square loop FSS

te

d

M

an

us

An equivalent circuit for the square loop FSS as proposed in [16] is shown in Fig. 9(b). The substrate with its characteristic impedance Zsub is sandwiched between the two FSS sections, each represented by a lumped inductance and capacitance in series. Z0 is the characteristic impedance of air. The values of L and C can be computed using Equations 1- 4 [17].

Ac ce p

Here L is the inductive reactance associated with L and C is the capacitive susceptance associated with C. G is a correction factor and ignored in the present calculations. The admittance of the shunt branch representing the FSS is then

and the magnitude of the transmission coefficient for a single layer FSS is given by

The resonance frequency can be obtained under the condition

For a single layer, square loop FSS with dimensions as given in Table II and with r = 2.7 (the mean of air and FR4), the calculated value of L is 5.2416-1, C is 440.929-1. With these values and equation (7), the calculated resonance comes out to be 6.24 GHz which matches well with the

Page 10 of 21

cr

ip t

simulated value of 6.27 GHz. Also, the transmission coefficient magnitude || calculated using (5&6) is shown superimposed on the simulated one in Fig. 9. The distance from the antenna at which the FSS screen is to be placed is decided from the reflection phase behavior. The distance should allow for constructive interference between the antenna radiation and the radiation reflected from the FSS. For this requirement to be met, a simple expression can be written as given in equation (8).

M

an

us

In the equation, is the reflection phase of the FSS, h is the distance at which the FSS is to be placed and β is the propagation constant of free space given by 2/. If we choose reflection phase to be 00, from equation (1), the optimal height can be shown to be h = /2. Hence, the optimal height can be taken equal to half of the wavelength at that frequency where the reflection phase of the FSS is zero. For the designed FSS, from the reflection phase characteristics shown in Fig. 9, the frequency at which the reflection phase is 00 is found to be 7.33 GHz. Hence, the optimal height is calculated as 21 mm.

Ac ce p

te

d

In Fig. 10(a), a photograph of the fabricated FSS screen is shown while the antenna along with the FSS screen mounted below (at optimal height of 21 mm) is shown in Fig. 10(b).

(a) (b) Fig. 10 (a). Photograph of the proposed FSS (b) setup with the slot antenna

Page 11 of 21

te

d

M

an

us

cr

ip t

A comparison of the simulated and measured reflection coefficients of the antenna without the FSS and with the double layer square loop FSS is shown in Fig. 11 and Fig. 12. Fig. 11 shows the comparison of simulated values whereas Fig. 12 shows the comparison of the measured values. It can be seen from Fig. 11, that the impedance bandwidth more or less remains the same after the application of the FSS screen. However, at lower frequencies (2 GHz to 3.5 GHz), the impedance matching slightly deteriorates in the presence of the FSS. In case of the measured results, however, the deterioration is not so evident. It is to be noted here that the operating region of the FSS is from 4 GHz to 11 GHz (Fig. 8) and higher impedance mismatch due to induced currents from reflected radiation with varying phase at other frequencies can be expected.

Ac ce p

Fig. 11. Comparison of simulated reflection coefficients without and with the FSS

Fig. 12. Comparison of measured reflection coefficients without and with the FSS

Page 12 of 21

Ac ce p

te

d

M

an

us

cr

ip t

The measured and simulated radiation patterns of the proposed slot antenna without the FSS are shown in Fig. 13. The radiation patterns are for the X-component of the gain (with reference to Figure 1) and plotted for select frequencies in the two principal planes (XZ, Phi = 00 and YZ, Phi = 900). From the figure, it is seen that the E-Plane (XZ plane) radiation patterns resemble dumb bell shape whereas the H-Plane patterns are quasi-omnidirectional in nature. There is a good agreement seen between the measured and simulated radiation patterns with the slight difference caused due to assembly misalignments. Also, most of the patterns seem to be reasonably stable with respect to frequency. The H-plane pattern at 6.0 GHz shows a pinch-off along the end fire directions ( =  900). This is because of the increased cross polarization at this frequency. The measured and simulated radiation patterns for the antenna with the FSS are shown in Fig. 14. As expected the radiation becomes more unidirectional in nature and the back lobes (along  = 1800) are considerably reduced. Also, the pinch-off seen at 6.0 GHz along the end-fire directions in case of the antenna without the FSS is not much evident in Fig. 14, more so in case of the measured result. This indicates a reduction in the cross polar component after the application of the FSS.

Fig. 13. Measured and simulated radiation patterns (antenna without FSS)

Page 13 of 21

ip t cr us an M

d

Fig. 14. Measured and simulated radiation patterns (antenna with FSS)

Ac ce p

te

A comparison between the measured peak gain (linear, total) for the antenna without the FSS and for the antenna with the FSS is shown in Fig. 15. The simulated peak gains for the two cases are also plotted alongside. For both the simulated and measured gains, an improvement with the FSS is clearly visible. The increase in the gain basically results from an increase in the directivity. On an average, the improvement in the peak gain is seen to be about 4 dB. It is also observed from the figure that the measured gain for both the cases remains low (quite less than the simulated values) up to 4.0 GHz. This is because of the poor characteristics (low gain) of the reference horn antenna (Schwarzbeck 4120C) used in the Antenna Measurement System in this range. The tendency for all the gains (measured, simulated, without and with the FSS) to increase with an increase in the frequency is attributed to the increased aperture area at higher frequencies. The maximum peak gain obtained with antenna after the application of the FSS screen is 11 dB (simulated) and 10 dB (measured). The simulated radiation efficiencies of the antenna without and with the FSS are shown in Fig. 16. There is approximately a 10% reduction in the efficiency in case of the antenna with the FSS and this is due to the additional losses taking place in the FSS screen and substrate. Nonetheless, the efficiency stays above 70% in the operating band.

Page 14 of 21

ip t cr us an

Ac ce p

te

d

M

Fig. 15. Measured and simulated peak gain of the proposed antenna without and with the FSS

Fig. 16. Simulated radiation efficiencies of the proposed antenna without and with the FSS

V. Modified Antenna and FSS Design

The FSS design presented in the previous section led to an enhancement in the antenna gain. However, with the FSS, the CP performance of the antenna is found to be affected. The axial ratio of the antenna with the FSS presented in Section-IV is shown in Fig. 17. For better comparison, the direction along which the axial ratio is shown is same as that for the antenna without the FSS i.e.,  = 730 and  = -890. It is seen that the axial ratio bandwidth has reduced with the FSS. This is because of the phase changes in the electric field components induced during reflection from the FSS and interference between the forward travelling and reflected waves. In another experiment, in

Page 15 of 21

te

d

M

an

us

cr

ip t

place of the FSS, a PEC is investigated for obtaining the desired objective. Three possibilities are considered; a PEC ground plane printed on the entire back side of the antenna substrate, partial ground plane on the antenna backside and PEC as a reflector placed 21 mm below the antenna. The simulated reflection coefficients and axial ratios (along  = 730 and  = -890) for all the cases are shown in Fig. 18. With a ground plane covering the entire backside of the substrate, it is seen that both the axial ratio and impedance matching have deteriorated. The antenna behaves almost like a microstrip patch antenna with a single resonance frequency. When the ground plane is partial (as for a ground backed CPW excitation), the return loss profile is almost similar to that of the original slot antenna with some increase in the impedance mismatch at lower frequencies. The axial ratio is disturbed and the CP bandwidth reduces. In the third case, when a PEC reflector is placed below the antenna, impedance matching gets affected at the lower end while CP gets disturbed at the upper end.

Ac ce p

Fig. 17. Axial ratio (along  = 730 and  = -890) of proposed antenna with FSS of Section-IV

Fig. 18. Simulated S11 and axial ratio (along  = 730 and  = -890) of proposed antenna with ground backing and with a PEC reflector

Page 16 of 21

us

cr

ip t

Hence, to simultaneously achieve the objective of high gain and wideband circular polarization, the antenna design is modified and a different type of FSS is designed. The schematic of the modified antenna and the FSS are shown in Fig. 19.

Fig. 19. (a) Modified antenna i.e., antenna – 2, (b) FSS-2, (c) unit cell dimensions of FSS-2

Ac ce p

te

d

M

an

As seen from the figure, in the modified antenna, the two corners of the slot to which no L-shape strips are attached are blended (rounded) by circular arcs of radii R1 = 11mm and R2 = 9 mm. In addition some of the parameters as outlined in Figure 1 have their values changed from those given in Table I. The new values for these parameters are; R = 5.9mm, W1 = 3.5mm, a = 14mm, b= 4.5 mm, g = 0.55mm and h = 0.9mm. The inclination of the slit on the circular disc is now 520 with vertical. The FSS-2 to be used along with the modified antenna is a patch type FSS with rectangular unit elements as proposed in [18] and printed on a FR4 substrate of thickness 3.2 mm. The simulated reflection coefficient, axial ratio and peak gain of the modified antenna (Antenna – 2) with FSS-2 placed below (at a distance of 24 mm) is shown in Fig. 20, Fig. 21 and Fig. 22 respectively. Also plotted in these figures are the characteristics of the original slot antenna. It can be seen from these figures that nearly same impedance and axial ratio bandwidths have been achieved for the modified antenna with FSS-2 when compared to the original antenna. As for the original antenna, the direction along which the axial ratio is shown is  = 730,  = -890. The gain enhancement achieved with FSS-2 is around 3 – 4 dBi.

Fig. 20. Return loss of modified antenna (antenna – 2) with FSS-2

Page 17 of 21

ip t cr

te

d

M

an

us

Fig. 21. Axial ratio (along  = 730,  = -890) for the modified antenna (antenna – 2) with FSS-2

Ac ce p

Fig. 22. Peak gain of modified antenna (antenna – 2) with FSS-2 Parametric Study of the FSS parameters The FSS discussed above is a patch type FSS with the patch parameters controlling the performance. The variation in the antenna performance with a change in the patch parameters is shown in Fig. 23, Fig. 24 and Fig. 25. The patch parameters varied are the length L (shown to be 15.5 mm in Fig. 19) and the width W (shown to be 2.9 mm in Fig. 19). Here, while varying these parameters, the periodicity of the FSS is kept constant. The performance indices checked are the reflection coefficient (Fig. 23), the axial ratio (Fig. 24) and the peak gain (Fig. 25). From Fig. 23, it is seen that when the patch dimensions are reduced (either by reducing L or W, the impedance matching at lower frequencies (2-4 GHz) improves. On the other hand, with reduced patch dimensions (consequently with reduced metallization), the gain also reduces as evident from Fig. 25. Finally, the variations in the axial ratio with changes in the patch dimensions as seen from Fig. 24 indicate the presence of optimal values for the best performance.

Page 18 of 21

ip t

an

us

cr

Fig. 23. Variations in reflection coefficient with FSS-2 patch length and width

Ac ce p

te

d

M

Fig. 24. Variations in axial ratio (along  = 730 and  = -890) with FSS-2 patch length and width

Fig. 25. Variations in peak gain with FSS-2 patch length and width

VI. Conclusion

In this paper, the design of a circularly polarized slot antenna is presented and the performance experimentally validated. The radiating element in the antenna is a rectangular slot which is excited using a stepped coplanar waveguide. The feed line is terminated on a circular disc shaped protrusion. For achieving circular polarization, inverted L-shaped strips are attached to the corners of the rectangular slot while a slant rectangular slit is cut on the circular disc. The measured axial ratio bandwidth is 2.35 GHz (40%, 4.75-7.1 GHz) embedded within the impedance bandwidth of 7.7 GHz (121%, 2.5-10.2 GHz). To improve the gain of the antenna, two frequency selective surface designs are presented. The first is a square loop designed at the center frequency of 7.33 GHz which improves the antenna gain by about 4 dB. The second is a patch type frequency selective surface which, besides enhancing the gain, retains the axial ratio bandwidth achieved

Page 19 of 21

without the frequency selective surface. The measured results wherever presented are found to be in good agreement with the simulated results.

ip t

Acknowledgement

cr

The authors are research fellows at DIAT (Deemed University), Pune, India and acknowledge the financial support provided for carrying out the research work.

References

Ac ce p

te

d

M

an

us

1. Lee KF, Luk KM. Microstrip patch antennas. Imperial college press. London. 2011. 2. Yahya R, Denidni TA. Design of a new dual-polarized ultra-wideband planar CPW fed antenna. In: 2011 IEEE International Symposium on Antennas and Propagation. p.1770-2. 3. Sze J-Y, Pan S-P. Design of CPW-fed circularly polarized slot antenna with a miniature configuration. IEEE Antennas Wireless Propag Lett 2011; 10; 1465-8. 4. Wang C-J, Chen C-H. CPW-fed stair-shaped slot antennas with circular polarization. IEEE Trans Antennas Propag 2009; 57(8); 2483-6. 5. Zhou S-W, Li P-H, Wang Y, Feng W-H, Liu Z-Q. A CPW-fed broadband circularly polarized regular-hexagonal slot antenna with L-shape monopole. IEEE Antennas Wireless Propag Lett 2011; 10; 1182-5. 6. Sze J-Y, Hsu C-IG, Chen Z-W, Chang C-C. Broadband CPW-fed circularly polarized square slot antenna with lightening-shaped feedline and inverted-L grounded strips,” IEEE Trans Antennas Propag 2010; 58(3); 973-7. 7. Chen Q, Zheng H-L, Quan T, Li X. Broadband CPW-fed circularly polarized antenna with equiangular tapered-shaped feedline for ultra-wideband applications. Progress in Electromagnetics Research C 2012; 26; 83-95. 8. Jeevanandham N, Nasimuddin, Agarwal K, Alphones A. Dual-band circularly polarized hexagonal-slot antenna. In: 42nd European Microwave Conference (EuMC). 2012, p.834-7. 9. Yang F, Samii YR. Electromagnetic band gap structures in antenna engineering. Cambridge university press. 2009. 10. Pirhadi A, Bahrami H, Nasri J. Wideband high directive aperture coupled microstrip antenna design by using a FSS superstrate layer. IEEE Trans Antennas Propag 2012; 60(4); 2101-6. 11. Kushwaha N, Kumar R. Design of slotted ground hexagonal microstrip patch antenna and gain improvement with FSS screen. Progress in Electromagnetics Research B 2013; 51; 177-99. 12. Ranga Y, Matekovits L, Weily AR, Esselle KP. A constant gain ultra-wideband antenna with a multi-layer frequency selective surface. Progress in Electromagnetics Research Letters 2013; 38; 119-25.

Page 20 of 21

13. Chen H-Y, Tao Y. Bandwidth enhancement of a U-slot patch antenna using dual-band frequency-selective surface with double rectangular ring elements. Microwave and Optical Technology Letters 2011; 53(7); 1547-1553.

ip t

14. Ranga Y, Matekovits L, Weily AR, Esselle KP. A low-profile dual-layer ultra-wideband frequency selective surface reflector. Microwave and Optical Technology Letters 2013; 55(6); 1223-7.

cr

15. Al-Joumayly, M, Behdad N. A new technique for design of low-profile, second-order, bandpass frequency selective surfaces. IEEE Trans Antennas Propag 2009; 57(2); 452-9.

an

us

16. Langley RJ, Parker EA. Equivalent-circuit model for arrays of square loops. Electron Lett 1982; 18; 294-6. 17. Chung Y-C, Lee K-W, Hong I-P, Lee M-G, Chun H-J, Yook J-G. Simple prediction of FSS radome transmission characteristics using an FSS equivalent circuit model. IEICE Electronics Express 2011; 8(2); 89-95.

Ac ce p

te

d

M

18. Karkkainen K, Stuchly M. Frequency selective surface as a polarization transformer. IEE Proceedings – Microwaves, Antennas and Propagation 2002; 149(5), 248-52.

Page 21 of 21