In this paper we describe a new freestanding periodic structure which can be used to provide low loss polarisation independent beamsplitting of two frequency bands when the filter is orientated at 45° incidence. The bandpass demultiplexer consists of one or more screens of closely packed nested annular slots which support different magnetic current modes in the inner and outer elements at resonance. These field distributions are obtained by choosing the diameter of the two slots and the position of the shorts in relation to the orientation of the incident electric vector. The design of a polarisation independent 2-layer Frequency Selective Surface which separates the bands 316.5-325.5GHz and 349.5-358.5GHz with > 20dB isolation and <1dB insertion loss is described. In order to validate the computer model a single layer beamsplitter, which is the basic building block of the cascaded screen structure, has been fabricated using a precision micromachining technique. The 30mm diameter periodic array was patterned in a 10µm thick silicon layer using the deep reactive ion etching technique, and finally the exposed surfaces were coated with a 1µm layer of electroplated copper. Quasi-optical swept frequency transmission coefficients which were measured over the range 250-360GHz in the TE and TM planes are shown to be in excellent agreement with the computed results. A concept for improving the robustness of the structure in order to provide space qualified beamsplitters for future polarimetric atmospheric sounders will be presented in the full paper.

Latest developments of wireless communication technologies have recently brought about a renewed need for the design of circularly polarized (CP) antennas. In particular, fast-growing RFID applications require compact, low-cost CP antennas to be used in various RFID read-write (R/W) systems. Conventional approach in designing CP planar printed dipole/monopole antennas uses two separate antenna elements fed by their own feed networks usually containing phase shifters or 90-degree hybrids, which contributes to the system design complexity. This paper presents the design and operational principle of a novel CP loop coupled dipole antenna (LCDA). The objective of this work has been to design an RFID R/W antenna having simple unobstructed layout and single feeding port yet demonstrating good CP performance.

The proposed design is presented in Fig. 1 showing the top view of a planar antenna printed on one side of a glass substrate. The antenna consists of two parts: a 2 mm-wide strip dipole element fed at its center by a conventional 50-Ohm balanced network, and pair of EM coupled rectangular loops being assembled by 1 mm-wide strips and placed in a dipole’s close vicinity as shown in Fig. 1. The idea is that in a far-field zone EM coupled loops would produce E-field (El) that is normal and equal in amplitude to the dipole radiation field (Ed) with 90-degree phase difference between them at a design frequency of CP operation. These design requirements have been achieved by carefully selecting the shape of rectangular loops and their placement with respect to the dipole element so that the key design parameters are the distance between the dipole and loop strips, the distance from loop center line to the dipole end point, and the loop length-to-width ratio. These parameters have been numerically optimized to realize the best CP at RFID operation frequency of 953 MHz. As a result, minimum CP axial ratio of 0.02 dB is obtained. Fig. 1 also presents the calculated 3D pattern of RHCP antenna gain at 953 MHz. LCDA shown in Fig. 1 produces RHCP E-field as observed along the z-axis. However, if one places coupled loops symmetrically with respect to the dipole element line LHCP radiation is easily obtained.

Another LCDA design that uses the principle described here includes placement of the antenna elements on both sides of a dielectric substrate, i.e. one strip of dipole element and one loop coupled to it printed on one side with an another pair of dipole element-coupled loop being placed on the opposite side of a substrate. EM simulations show that such an antenna has almost the same CP characteristics as compared to one-sided LCDA. Several antenna prototypes based on the design presented here have been made and measured. Measurement results have confirmed the CP characteristics of the proposed LCDA. It should be noted that besides RFID R/W design the presented antenna is an attractive candidate for various wireless communication applications requiring CP operation, such as GPS and ETC.

In recent years, there has been increased interest in the use of electromagnetic bandgap (EBG) structures as a ground plane for low profile wideband antennas. Fig 1 shows the antenna configuration. To design of the EBG structures some parameters such as thickness and permittivity of substrate, width of patches and cell size must be considered. If the substrate specifications are supposed to be constant, then there will be many possibilities for the design the EBG structures. This means that many EBG structures can be designed with the same resonance frequencies but different element and cell sizes. So, to find the best design, three different EBGs with the same substrates and different patch sizes which result the same resonance frequencies were chosen. Then a diamond dipole antenna was located above the EBG ground plane and the resulting antenna was simulated by HFSS software. Fig. 2 and 3 shows the reflection phase and return loss of the antenna on the EBGs respectively. As can be seen, by increasing the width of patches the reflection phase after the resonance frequency drops more rapidly. Consequently, for larger elements the antenna S11 at higher frequency degrades. The reflection phases of all the EBGs below resonance are approximately the same but still affect on the return loss. The radiation pattern of the antenna on the EBGs with bigger elements have better performance.

Next generation of wireless communications will require systems with broadband capabilities in small robust antennas. Compact Multiple-Input Multiple-Output (MIMO) antenna is a suitable candidate for fulfilling mobile applications demands for high speed and high quality transmission. However, closely spaced antennas suffer from mutual coupling and highly correlated signals. Different methods are commonly employed to reduce these effects, consisting basically in adding parasitic or circuit decoupling elements.

In this paper, a miniature multi-layered broadband four-element antenna with an optimized decoupling technique is presented. The antenna structure (Fig.1), which has been implemented in a reduced area of , operates at the wireless frequency band of 2.45 GHz. Each of the four radiating elements consists of two stacked patches over a ground plane. Four decoupling bridges have been placed between those elements in order to decrease mutual coupling, maximizing the capacity in a complete RF system channel model.

In order to have realistic capacity figures, a channel model has been developed considering the complete RF transmission chain, including the signal source (transmitter), transmit antennas, physical channel, receive antennas, and signal drain (receiver) and taking into account all mutual coupling factors. The capacity can then be expressed as:

From the original 4-element antenna, and applying the decoupling bridges first and the waterfilling technique afterwards, additional successively improvements of 25% have been obtained.

Details of the four-element decoupled antenna and comparison between numerical and experimental results for the system capacity for different channel models will be presented in the full paper.

References:

[1] L. Jofre, B.A. Cetiner, F. De Flaviis, "Miniature Multi-element Antenna for Wireless Communications", IEEE Trans. on Antenn. and Propag., vol. 50, n. 5, may 2002, pp. 658-669.

[2] Kyeong-Sik Min, Dong-Jin Kim, Young-Min Moon, "Improved MIMO Antenna by Mutual Coupling Suppression between Elements", 8th European Conference on Wireless Technology 2005, Paris.

Fig. 1: Four-element antenna design	Fig. 2: Results obtained for S11, S21,S31

Recently, Federal Communication Commission (FCC) adopted ultra-wideband (UWB) for short-range peer-to-peer ultra-fast communications. This allocation has excited antenna designers for challenging design of low cost ultra-wideband antennas. Several antenna monopole-like UWB antennas have been proposed to support UWB communications. In particular, printed antennas have been receiving considerable attention due to their low cost, lightweight, and ease of construct. Their flat structures facilitate using of them in PC devices for personal computers or inside mobile phone handsets. However, among planar antennas with various shapes, circular and square antennas have been shown to suffer from relatively small bandwidth. Elliptic planar antennas, on the other hand, exhibit superior broadband performances. It has been shown that they have a return loss -10 dB or better for minor axis larger than 0.2ë. However, there are some challenging problems in the design of these antennas. One of them is the feeding network. In most available designs, the distance between the radiating element and ground plane is very small. Therefore a feeding mechanism is required in order to properly feed the antenna. In addition, since the performance of the antenna degrades significantly in the vicinity of PEC circuit boards, it can not be located close to PEC boards. Therefore, for most applications, the antenna should be placed outside the device and hence, the size of the antenna should be as compact as possible.

The objective of this paper is to design a planar miniaturized elliptic-card antenna to cover 3-11 GHz band. This antenna consists of an elliptical radiating element and a ground plane both lying in the same plane. Since the distance between ellipse and ground plane is very small (0.375 mm), in order to feed the antenna a microstrip feed line is located on the other side of the substrate and it is connected to the elliptical element by a via. The line is fed in the low current region of the antenna to prevent the perturbation of the ground plane current. Simulation results of this structure show a VSWR less than 2 over the span of interested frequency band (3-11 GHz). The antenna with the feed system was built and measured. The antenna was printed on a Duruid substrate with ĺr=2.33 and thickness of 0.508 mm. Measured VSWR results verified the ultra-wideband behavior of the antenna. The simple and small size structure of this antenna introduces it as a very attractive candidate for UWB wireless communications.

Two examples of reconfigurable leaky wave antennas made of printed transmission structures with electronic devices will be reviewed from the mode switching point of view. The first example is switching between the microstrip patch antenna and a leaky wave antenna made of a higher order microstrip line mode. Another example is a leaky wave antenna made of the composite right/left hand (CRLH) metamaterial configurations. By way of bias control of varactor diodes in each unit cell of the antenna, the dispersion characteristics are varied for beam shape and gain control.

Multiple Input Multiple Output (MIMO) wireless communication systems have been a focus of interest, due to their ability to increase the capacity in rich scattering environments by using multi-element antenna arrays both at the transmitter and the receiver side. One of the fundamental issues concerning MIMO systems is the choice of the array type and configuration. Although printed arrays are advantageous over other antenna types for their low cost, light weight and conformability to the mounting surface, they are not investigated adequately in terms of MIMO performance.

In this paper, MIMO performance of planar and conformal printed arrays is analyzed through the MIMO channel capacity. The effects of mutual interactions among the array elements through space and surface waves are included into the channel matrix using a full-wave hybrid Method of Moments (MoM)/Green's function technique in the spatial domain. Regarding the planar arrays, two different Green's function representations are used. Basically, the efficient integral representation of the planar microstrip dyadic Green's function is used around the source region (diagonal and nearly diagonal terms of the impedance matrix), and high-frequency based asymptotic closed-form representation of the grounded dielectric slab Green's function is used away from the source region.

On the other hand, in order to compute MoM matrix elements of the conformal arrays, three different spatial-domain Green's function representations, each accurate and computationally efficient in a given region of space, are used in conjunction with a switching algorithm. These Green's function representations are as follows: (i) The planar representation which is valid when the field is evaluated in the vicinity of the source. It is used based on the assumption that for electrically large material coated circular cylinders and small separations between the source and field points, the surface can be treated as locally flat. Hence, an efficient integral representation of the planar microstrip dyadic Green's function is used for the self term evaluations of the impedance matrix. (ii) The steepest descent path (SDP) representation of the dyadic Green's function, which is used away from both the paraxial (nearly axial) and the source regions. This representation tends to become more efficient and accurate as the separation between the source and field points increases. (iii) The paraxial spatial domain representation of the dyadic Green's function, which complements the SDP representation along the paraxial region.

In order to assess the performances of planar and conformal arrays, we have compared the MIMO capacities to that of dipole arrays of thin wires. The geometric parameters of dipole arrays and printed arrays are chosen to be the same. As expected, the preliminary results show better performance of printed arrays. More results will be presented for the printed arrays involving the array size, dielectric properties and array geometry.

An accurate mathematical and numerical analysis of both a Lüneburg lens and a slotfed sphericalcircular printed antenna with a sphericalcircular ground conductor is fully presented. The entire analysis is done in terms of the spherical vector wave function expansions in each partial domain. The feed is modeled by an horizontal magnetic dipole used as a forcing current. Applying the boundary conditions lead to the overall scattered electromagnetic field at any spatial position. The problem is cast into a coupled set of the dualseries equations for the expansion coefficients, and then to an infinitematrix equation having favorable features by expressing the currents into the Fourier domain. This is achieved by following the Method of Analytical Regularization, which is based here on the explicit inversion of the static part of the dualseries equations thanks to the Abel integral equation properties. Such a procedure leads to a guaranteed convergence and controlled accuracy of computations in accordance with the Fredholm theorem of the Fredholm equations of the second kind. Because of its semi analytical nature, the implemented algorithm is accurate and very low in both CPU time and memory capacity consumption. This highlights several properties of the antenna lens structure. Indeed, the modes associated with each and every part of the entire structure are extracted and their effects analyzed, mainly in terms of both the metallic cap sizes and the number of layers composing the lens: generalized cavitymethod patch modes, ground effects, traveling modes in each dielectric layer of the lens and in the coating, curvature influence. These are translated in terms of classical antenna quantities: directivity, far field pattern, current. The presented theory is backedup by some comparisons with measurements.

The researchers of the Electronics, Antennas and Telecommunications Laboratory (LEAT) of the University of Nice started to focus on printed antennas in 1985.

Without any commercial software available in the eighties, the first goal of each lab interested in printed antennas was to implement accurate modelling techniques. In 1987, several rectangular patches printed on a dielectric substrate were successfully analyzed by resolving the integral equations in the spectral domain (moment method). This method was further developed to analyze coaxial and microstrip line fed elements as well as sandwiched patches between two dielectric layers (1988). Simultaneously, the same researchers enhanced a simple cavity method by taking into account the coaxial feed of the antenna. Lastly, an efficient numerical technique, the Transmission Line Matrix (TLM) traditionally used for microstrip lines and waveguides, was applied to radiation’s problems (1990). At the beginning of the nineties, with these home-made tools, the researchers of the LEAT were able to analyze and investigate new ideas of printed structures with arbitrary shapes. Numerous passive and active antennas were designed and fabricated in the following years.

It is the purpose of this paper to give an overview of the most interesting structures designed at the LEAT during more than 20 years.

We will particularly focus on:

Numerical full-wave methods applied to analysis of periodic structures ask for fast and accurate evaluation of the periodic Green's functions. The Ewald transformation, originally developed by Ewald [1] has been advantageously used in the efficient evaluation of Green's function of an infinite periodic phased array of line sources (2-D structures with 1-D periodicity) [2], 2-D structures with 2-D periodicity [3], for three-dimensional (3-D) problems with 2-D orthogonal lattices [4] as well as for rectangular cavities (3-D problem with 3-D orthogonal lattice) [5]. Its application in evaluating GFs for planar multilayered media with 2-D periodicity is treated in [6]. In this paper, we present the Ewald sums for 3-D Helmholtz equation with 2-D skewed lattices. We develop and present for the first time to our knowledge, the expression for the optimal value of the splitting parameter for 2-D skewed lattices and show that it reduces to the formula reported in [4] and [7] for the case of orthogonal 2-D lattices. Moreover, we derive the gradient of the scalar potential GFs and address the extraction of singularity for both vector/scalar potential GF and their curl/gradient. This is a mandatory step when solving periodic printed structures (infinite arrays) with a mixed potential integral equation (MPIE). Several numerical implementation issues are also discussed leading to further enhancement in computational speed, accuracy, and numerical stability. We apply the developed technique to an MPIE analysis of a wide variety of periodic structures ranging from frequencyselective surfaces and quasi-optical filters to electronic bandgap materials with general triangular lattices and arbitrary plane wave excitations.

REFERENCES

[1] P. P. Ewald, "Die Berechnung Optischer und Elektrostatischer Gitterpotentiale," Ann. Phys., vol. 64, pp. 253–287, 1921.
[2] F. Capolino, D. R. Wilton, and W. A. Johnson, "Efficient computation of the 2-D Green's function for 1-D periodic structures using the Ewald method," IEEE Trans. Antennas Propagat., vol. 53, no. 9, pp. 2977–2984, Sept. 2005.
[3] A. W. Mathis and A. F. Peterson, "Efficient electromagnetic analysis of a doubly infinite array of rectangular apertures," IEEE Trans. Microwave Theory Tech., vol. 46, no. 1, pp. 46-54, Jan. 1998.
[4] K. E. Jordan, G. R. Richter, and P. Sheng, "An efficient numerical evaluation of the Green’s function for the Helmholtz operator on periodic structures," J. Comput. Phys., vol. 63, pp. 222-235, 1986.
[5] M.-J. Park, J. Park, and S. Nam, "Efficient calculation of the Green's function for the rectangular cavity," IEEE
Microwave Guided Wave Lett., vol. 8, no. 3, pp. 124-126,Mar. 1998.
[6] M.-J. Park and S. Nam, "Rapid calculation of the Green's function in the shielded planar structures," IEEEMicrowave Guided Wave Lett., vol. 7, no. 10, pp. 326-328, Oct. 1997.
[7] A. Kustepeli and A. Q. Martin, "On the splitting parameter in the Ewald method," IEEE Microwave Guided Wave Lett., vol. 10, no. 5, pp. 168-170,May 2000.