Electromagnetic band gap (EBG) structures have been used in the design of high-gain planar antennas. Metallic arrays have been placed in front of simple radiating sources in ground plane to increase their directivity and thus produce high gain antenna designs [1-3]. Dielectric EBG materials have also been used as superstrates for gain enhancement of dipole and patch antennas [4-5]. Both of the above structures rely on the formation of a resonant cavity between a highly reflective structure and the ground plane. The cavity resonance increases the directivity of simple radiating sources, however it narrows significantly the operational bandwidth. A new method for obtaining broadband resonant cavities maintaining the high-gain performance is presented here. The design is based on a double-layer metallodielectric array structure with dissimilar array element size. The MEBG array exhibits a characteristic reflection phase response which satisfies the resonance condition of the cavity at a wide range of frequencies. Broadband and highly-directive antenna designs have been obtained by placing the MEBG array over a waveguide-fed slot in a ground plane. Similar designs are also being investigated for microstrip patch fed antennas.

References

[1] G. V. Trentini, "Partially reflecting sheet arrays," IRE Trans. Antennas Propag., AP-4, pp. 666-671, 1956
[2] J. R James, S. J. A Kinany, P.D Peel, and G. Andrasic, "Leaky-wave multiple dichroic beamformers," Electron. Lett., 1989, 25, pp. 1209-121
[3] A. P. Feresidis and J. C. Vardaxoglou, "High gain planar antenna using optimised partially reflective surfaces," IEE Proc. Microw. Antennas Propag., vol. 148, no. 6, pp. 345-350, Dec. 2001.
[4] H. Y. Yang and N. G. Alexopoulos, "Gain enhancement methods for printed circuit antennas through multiple superstrates," IEEE Trans. Antennas Propag., vol. 35, no 7, pp. 860-863, 1987
[5] C. Cheype, C. Serier, M. Thevenot, T. Monediere, A. Reineix, and B. Jecko, "An electromagnetic band gap resonator antenna," IEEE Trans. Antennas Propag., vol. 50, no 9, pp. 1285-1290, 2002

In this paper, a new method to increase the E-plane directivity of single dipole antennas by incorporating cylindrical metallo-dielectric electromagnetic bandgap (CMEBG) surfaces. Initially, the proposed structure is analysed for infinite CMEBG array length which is centre-fed by a simple radiating source. The analysis yields insight for the design and realisation of an omni-directional H-plane CMEGB antenna with enhanced E-plane directivity. The use of aperture arrays is employed to achieve significantly smaller unit cell. Furthermore, a reduction of the grating lobes that conducting arrays suffer from is achieved and presented. A comparative study between aperture and conducting CMEBG arrays illustrates what are the advantages and disadvantages of each structure. The Cylindrical MEBG arrays behave as Partially Reflecting Surfaces (PRSs) forming a cylindrical resonant cavity. A balanced dipole antenna is used as the feeder at the centre of the cavity.

The directivity enhancement is demonstrated with simulated and measured results for an antenna prototype operating at 2.4 GHz. A maximum directivity improvement of about four times is achieved.

Fig 1: The proposed CMEBG antenna using periodic aperture array.

Fig 2. The proposed CMEBG antenna using periodic conducting array.

Fig 3. 3D radiation patterns with and without the CMEBG array.

The enhancement of directivity by using leaky wave antennas has been a subject of study in the last forty-fifty years and various geometries have been proposed. The enhanced directivity is based on the excitation by a single radiator of a leaky wave along the antenna geometry. When the leaky wave has a small propagation constant the antenna radiates a narrow beam at broadside. Various configurations have been considered to enhance the directivity at broadside, such as single or multi dielectric layers to form a cavity resonator often called Fabry-Perot cavity (FPC). A source inside the FPC excites the leaky mode, so that a large radiating aperture just outside (above) the cavity is sustained, resulting in a large gain. Antenna directive features are selected by properly modulating the leaky mode properties, i.e., the FPC and PRS arrangement. The PRS can be effectively realized printing a periodic metallic geometry on a thin dielectric substrate, thus creating a Frequency Selective Surface (FSS), for which a wide literature is available to guide the design and to obtain specific features. When a planar FSS is suspended over a planar patch antenna, a FPC is created and a simple and low cost leaky wave antenna is assembled in planar technology. Since in the FPC arrangement, the directivity enhancement is obtain to the detriment of gain bandwidth, the design of sparse arrays inside the FPC is proposed to achieve high directivity still preserving the needed operating bandwidth. The thinning of the array results in a simpler structure with fewer elements, a simpler feeding network and a lower coupling between elements. Furthermore, the empty space available between sparse elements can be utilized to accommodate active RF circuitry or to interleave two different antennas, as in a dual feed arrangement.

In this paper, a design procedure for the optimization of the patch antenna inside the FPC will be presented. In particular the patch design and the FPC size optimization will be presented by resorting to an approximate model of the PRS consisting in a high permeability or permittivity thin superstrate layer, that allows the use of commercial planar antenna simulation and design tools. Note than in this case the layer is not a quarter-wavelength thick, as usual for these kind of FPC antennas. In a second step the PRS is designed resorting to well assessed FSS design criterions and software tools, which assume an infinite periodic structure and provide the same reflection and transmission characteristics of the thin homogeneous dense layer. By using an appropriate aperture size estimation rule, the actual truncated (periodic) PRS is finally inserted into the full wave model. Since this last step requires a considerable numerical effort, the previous intermediate steps are very important to speed up the design process by providing a nearly optimized antenna structure which only needs a fine refinement. The same approach can be used to provide guidelines for the design of thinned arrays of patches under a FSS. Here the numerical effort of the entire array antenna is challenging and the possibility of developing the design with the intermediate simpler steps becomes crucial.

A design example will be presented showing a FPC dual polarized antenna that comprises two interleaved thinned array of patches, one for each polarization, under a FSS.

Planar antennas are key elements in microwave and millimeter wave systems for many applications. Coplanar waveguide fed antennas are very attractive in term of compactness and performances. The fabrication of these types of antennae through collective micromachining techniques, available with GaAs and silicon technologies, will play a major role in the future.

In recent years, the micromachining technology has been proposed for the fabrication of high performance millimetre wave circuits on very thin dielectric membranes, mostly for silicon substrates. Micromachining of GaAs is an exciting less explored alternative for the RF-MEMS field. It is very interesting due to the easy monolithically integration on the same chip of micromachined passive circuit elements with active devices.

The Yagi-Uda antenna topology exhibits attractive radiation performances. It is a traveling-wave structure that, as the number of elements increases, has improved directivity, gain and front-to-back ratio. If the antenna is fabricated on a very thin dielectric or semi-insulating semiconductor membrane, quasi free space operating conditions are available for the antenna. This approach takes advantages of the experience accumulated for Yagi-Uda antennas for HF through UHF frequency range. Membrane supported Yagi-Uda antennas for an operating frequency of 45 GHz have been reported in [1].

This paper presents new developments of membrane-supported Yagi-Uda antenna structures for the 60 GHz frequency range (fabricated using GaAs micromachining) and for 77 GHz operating frequency (fabricated using silicon micromachining). Photos of the two types of the Yagi-Uda antenna structures are shown in Fig.1 and Fig.2, respectively.

The design and manufacturing are briefly described. The experimental set-up and characterization technique are explained in detail. The measurements are performed "on wafer" for a wide frequency range. The antenna structures gain as well as the radiation patterns will be presented in the final paper.

Acknowledgments

The authors acknowledge the support of the European Commission through the European Project 507352 "AMICOM".

References

[1] D. Neculoiu, P. Pons, L. Bary, M. Saadaoui, D. Vasilache, K. Grenier, D. Dubuc, A. Müller and R. Plana "Membrane Supported Yagi-Uda Antennae for Millimeter-Wave Applications", IEE Microwave, Antennae and Propagation, vol.151, No.4, pp.311-314, 2004

Fig. 1. The photo of the fabricated 60 GHz Yagi-Uda antenna structure

Fig. 2. The photo of the fabricated 77 GHz Yagi-Uda antenna structure

Many embedded communication systems need low-profile directive radiating structures. In this paper, we propose, to describe and validate a design procedure for the synthesis of EBG (Electromagnetic Band Gap) resonator antennas combined with planar arrays of sources. This methodology allows the optimum design of the antenna array and the EBG resonator in order to comply with given target specifications (e.g. broadside directive radiation pattern template). In particular, the resonator is designed to increase drastically the directivity of the feed as well as to suppress the grating lobes thanks to its spatial filtering capabilities. Compared to conventional antenna arrays, the use of EBG resonators lightens the design of feed networks and reduces their complexity and losses. In addition, when excited by an array of sources, the radiation bandwidth of this kind of antennas is enhanced significantly compared to conventional EBG resonator antennas excited by a single source.

The typical geometry of a metallic EBG resonator antenna excited by a patch antenna array is illustrated in Fig. 1a: the Fabry-Perot (FP) cavity is composed of a ground plane (usually, that of the feeding sources) and an upper PRS (partially reflecting surface); the planar array is embedded inside the cavity; the inter-element spacing between the radiating elements is typically larger than one wavelength in free space. Such a solution allows using a low-Q (wide band) EBG, and generating highly directive beams thanks to the spatially-distributed feeding of the resonator.

A reflective spiraphase-type phased array based on split metal rings with p-i-n diode switches is analyzed. This array can be considered as an array of metal ring reflectors arranged in a equilateral triangular grid with a period b and situated at a distance d over a metal screen (Fig. 1 a). A single element of the array is shown in Fig. 1 b. This element contains a split ring 1 and four pairs of the p-i-n diodes 2-6, 3-7, 4-8 and 5-9. At each moment of time one pair of diodes is switched off and another three pairs of diodes is switched on. Due to the different states of the diode pairs, the reflective element provides a differential phase shift of 180 degrees between two reflected linearly-polarized waves with polarization planes parallel to the axis u and v, respectively. Switching of the diode pairs leads to the electronic simulation of the mechanical rotation of the element on the angles 0, 45, 90 or 135 degrees. Thus, the element introduces additional phase shifts of 0, 90, 180 or 270 degrees into the reflected circularly polarized wave (CPW).

A developed full-wave mathematical model was used to optimize the array element for Ku band and to simulate the array characteristics. As a result, the geometry of the array was determined as follows: the triangular lattice size b is equal to 10 mm, the inner radius of the metal ring r1 is 2.90mm, the outer radius of the metal ring r2 is 3.05mm. A dielectric substrate of relative permittivity of 2.2 and thickness of 0.127mm is considered in the simulation. The distance d between the substrate and the metal screen is 3.5 mm. The parameters of HPND 4005 p-i-n diodes are used in the simulations.

The results of the simulation prove that the reflective array based on split metal rings ensure the efficient control of the reflection angle in a frequency band from 12 to 18 GHz. The incident CPW can be redirected in the directions determined by the elevation angles up to 50° with conversion loss less than 2.7 dB. The calculated conversion loss includes the dissipative loss in the p-i-n diodes and the conversion loss due to the phase errors that introduce digital phase shifter. For the case of the optimized two-bit spiraphase-type phase shifter with HPND 4005 diodes the minimum-possible insertion loss is 0.50 dB, meanwhile the conversion loss due to the phase errors that introduces a digital two-bit phase shifter is 0.91 dB.

The mast of a marine ship houses antenna systems operating at different frequency ranges, each one located in a separate compartment. The walls of these compartments can host a Frequency Selective Surface (FSS) designed to be transparent in the frequency band of the corresponding antenna and reflecting in the backscatter direction outside that band to reduce the antenna Radar Cross Section.

In this contribution, we focus on the design of an FSS for the compartment that houses telecommunication antennas. The FSS should guarantee a pass-band between 100 and 400 MHz, with maximum reflected power of -15dB, for elevation angles between 0° and 60° and azimuth angles between 0° and 90°. The FSS should reflect the incoming signals in the band 3-18 GHz, where typical airborne and shipborne radars are operational, for every azimuth angle. Moreover, the elevation angles of interest for the considered application are within the range 0°-30°.

Because of the large band required for this application, a suitable geometry is a type-4 gangbuster [1], with dipoles designed to resonate at 30 GHz and interelement separation of few millimetres, printed on duroid layers. A single gangbuster surface works only for the case of electric field parallel to the dipoles. To handle arbitrary polarizations, a super dense gangbuster has been considered, consisting of two equal surfaces cascaded at a small distance and twisted of 90° with respect to each other.

The FSS performances in terms of reflection and transmission coefficient for plane wave incidence need to be characterized at two completely different frequency ranges. While measurement techniques for the S- and X-bands are well known and available, measurements at l00-400 MHz pose some challenges. In this case, the FSS is always in the near field of the radiating antenna and, since the target antenna is omnidirectional, the diffraction from the edges of the FSS panel would have a prominent effect on the measured performances.

In the full paper, the design requirements, the measurement techniques applied for the two frequency bands as well as the trade-off that led to the achieved performances will be described.

[1] S. W. Schneider and B. A. Munk, "The scattering properties of "Super Dense" arrays of dipoles", IEEE Trans. Antennas Propagat. Vol. 42, n. 4, April 1994.

In this paper, a new planar Metamaterial based on dipoles and wires is presented. The unit cell is formed by three layers, two layers of dipoles and one layer with a wire. Due to the close proximity of the dipoles, currents flow through the gap between dipoles creating a loop which produces an effect similar to the one of a Split Ring Resonator (SRR) (µ < 0). By adding the continuous wires to the unit cell, which produce the negative permittivity (ε < 0), the negative refraction index is achieved.

The effect of each parameter of the dipoles and the wire in the resonant frequency of the cell has been analyzed in detail; distance between dipoles (h), distance from the dipoles to the wire (d_y), gap between dipoles (d_x), width of the dipoles and wire (d), thickness of the metallization (t), length of the dipoles (L) (see Fig. 1).

Figure 1. Configuration of the unit cell. (a) 1 layer of dipoles. (b) 2 layers of dipoles. (c) Whole cell: 2 layers of dipoles + wire. (d) Dipole antenna + meta-surface.

Due to the resonant behaviour of the cell, pass band and stop band of frequencies are observed. The printed dipoles reflect the power at their resonant frequency, being transparent out this frequency. However, the complete unit cell (dipoles + wire) presents a pass band due to the interaction between dipoles and wires. At that frequency, all the elements are resonating.

The unit cell has been analyzed step by step (1 layer of dipoles, 2 layers of dipoles and 2 layers of dipoles + wire) by means of their dispersion diagrams and by their transmission responses as function of the frequency. An infinite periodic structure has been studied by applying periodic boundary conditions (PBC) to the unit cell.

Because of the planar characteristic of the unit cell, it can be used as cover meta-surface of planar antennas with a small increase in the thickness of the whole configuration and with an enhancement of the gain. Simulations of a dipole antenna tuned to the pass band frequency of the meta-surface have been carried out, proving the improvement in the radiation performances of the whole configuration (dipole antenna + meta-surface).

Recently, there has been great interest in microstrip reflectarrays in general, and particularly in reconfigurable
microstrip reflectarrays, since they are planar, lightweight and potentially low cost. There have been attempts to
realize reconfigurable reflectarrays using semiconductor technology (PIN or varactor diodes) as well as MEMS.
We are proposing an alternative approach, that makes use of the dielectric anisotropy of liquid crystal to tune the
reflection phase of the reflectarray elementary cells. The unit-cells consist in principle of a superstrate, on which
one or stacked metallic patches are printed. A spacer between this superstrate and the ground plane provides a
thin cavity to be filled with the LC (Fig. 1).

Fig.1: Schematic of LC-reflectarray unit cell with simple patch

Due to potential applications and realization aspects, the mm-wave range (35 and 77 GHz respectively) was
chosen to demonstrate the concept. Different unit cells have been designed and manufactured. Simple patches
printed on one side of the substrate and stacked patches on both sides of the substrate have been investigated,
as well as liquid crystal cavities with 120 µm and 50 µm thickness. A thin mechanically rubbed polyimide film on
both sides of the cavity is used to provide a preorientation of the LC molecules. The DC-voltage applied between
patch and ground controls the LC permittivity, thus enabling phase adjustment.

The unit cells were measured in a waveguide simulator: a WR24 standard waveguide has been terminated with
two 35 GHz unit cells, and a WR10 waveguide, flared at the end so as to accommodate the unit cell, was
terminated with a 77 GHz unit cell. The complex reflection coefficients are measured each time.

(a) Stacked patches, 120 µm cavity, @35 GHz

(b) Simple patch, 50 µm cavity, @77 GHz

Fig.2: Measured results for 35 GHz and 77 GHz reflectarray unit cells

In Fig. 2 two characteristics are presented exemplarly: one for a 35 GHz unit cell with stacked patches and
120 µm cavity thickness, the other for a 77 GHz unit cell with a simple patch and 50 µm cavity thickness. The
phase variation over the control voltage is 300° (measured at 33 GHz due to manufacturing tolerances) in case of
the 35 GHz unit cell, and 280° in case of the 77 GHz unit cell. These values, although lower than the 360°
optimum, would be enough for a working full scale reflectarray. The highest return loss is in both cases around
-7 to -8 dB, whereas one can see in Fig. 2 (b), that in case of the 77 GHz, the losses stay around -4 dB even
outside resonance. This could be due to the specific measurememt setup, that allows part of the power to radiate,
an amount that we estimated at about 3 dB. One must however keep in mind, that the average losses on a large
array are lower than the highest losses at resonance.

In parallel, 35 GHz fixed-beam reflectarrays have been successfully realized, and after having demonstrated the
functionality of the tunable unit cells in principle, a 35 GHz reflectarray with reconfigurable main beam in one
dimension has been realised. Measurements thereof are now ongoing.

In this work the one-dimentional photonic crystals containing bianisotropic, backward wave (left-handed) [1], and usual isotropic layers that have no stop bands are considered.

For the first time it is shown analytically that rearrangement of layers within a period does not affect on bandgap of the crystal. In spite of a photonic crystal is a stratified meduim and the multireflection is an obligatory phenomenon for this, it also is shown that it is possible to construct the structures without stop band for any frequency. Mathematically it means that the eigennumber modules of the translation matrix [2] for a period will be equal to one for any wave frequency. It is obtained the rigorous analytical proof and given physical explanation that the following types of the crystals have no stop bands.

Type 1. Let us consider the two bianisotropic layers (Fig.1a) with the same constitutive parameters and with the different orientations of the bianisotropy axes ₂=-₁, β₂=β₁. The photonic crystal each period of which contains two such layers has no stop bands.

Type 2. A photonic crystal with double-layered period when the constitutive parameters of the layers are the same and the directions of the bianisotropy axes are opposite has no stop band.

Type 3. A photonic crystal with double-layered period (Fig1a) when the constitutive parameters of the layers are the same and the directions of the bianisotropy axes are defined by the angles ₂=₁, β₂=-β₁. in Fig 1a has no stop band.

Type 4. A photonic crystal with double-layered period (Fig1b) when the constitutive parameters of the layers are the same and the directions of the bianisotropy axes are defined by the angles ₂=-₁, β₂=-β₁. in Fig 1b has no stop band.

Tipe 5. A PC containing a bianisotropic (anisotropic, isotropic) layer, a multilayered isotropic structure, and a multilayered backward wave (BW) structure described by the matrix (Fig.1c) has no stop bands. Taking into account the property of rearrangement layers of periodic structures [3] it is obvious that the arrangement of layers within a period shown in Fig.1c can be arbitrary.

Such crystals can be constructed by using artificial materials and as result of twining of semiconductor crystals [4]. For example, these can be Si and Ge where the layer thickness is about 20nm - 20000 nm.

Such photonic crystals can be used in the whole of frequency domain, for example, for rotation of a polarization plane.

Reference
[1] I.S.Nefedov, S.A.Tretyakov 'Photonic band gap structure containing metamaterial with negative permittivity and permeability' Phys. Rev. E, vol. 66, 2002.
[2] E. Wolf, M. Born Principles of Optics, 3rd ed. Pergamon Press, Inc., New York, 19.
[3] K. A. Vytovtov 'Investigation of isotropic stratified media' J.Opt.Soc.Am.A. vol. 22 no. 4, pp. 689-696, 2005.
[4] M. S. Spector, S. K. Prasad, B. T. Weslowski, R. D. Kamien, J.V. Selinger, B.R. Ratna, R. Shashidar Chiral Twisting of a Smectic-A Liquid Crystal Phys. Rev. E 61 (2000) 3977-3995