Wideband and multiband antennas are vital to many applications ranging from handheld terminals, basestations, sensors, aircraft, satellites etc. As the number of services and functions increase, stringent constraints with respect to available space, mass limitations and the need to reduce cost dictate that several systems or functions use a single antenna. Some applications like ultra wideband communications and radar use unprecedented wide frequency bands and have demanding requirements, e.g. on antenna dispersion.

The EU 6th Framework Network of Excellence ACE - Antenna Centre of Excellence - includes wideband and multiband antennas as one of its five vertical joint research activities. The activity is divided into three work packages: WP 2.3-1 on wideband and multiband radiators dealt with in this paper, WP 2.3-2 on reflector surface models dealt with in a parallel paper, and WP 2.3-3 on antennas for ground or surface probing radars including medical imaging. The radiator work package focuses on comparatively large radiating elements including reflector antennas. Cooperation takes place with the activities A2.1 on mm and submm wave antennas, A2.2 on small antennas and A2.4 on arrays. There is a wide variety of wideband, multiband and ultra wideband antennas needed for existing applications that continue to develop as well as for emerging applications, and there is a wide range of expertise in these fields across Europe. Our main objective is to gather and integrate this expertise. To increase the relevance of the work, ESA/ESTEC participates in an advisory role. Strengthening the cooperation with colleagues in the new EU member states and improving gender policies have also high priorities.

The paper gives an overview of our wideband, multiband and ultra wideband antenna activities, describes our current results and some future plans, and relates the work to other ACE activities. Deliverables of the activity including a wideband antenna database are available from the ACE Community at http://www.antennasvce.org/.

Ultra wideband (UWB) antennas has made rapid advances in the past few years, receiving increased attention in the UWB radio and radar system applications. It is well known that self-complementary planar antennas show a constant input impedance which is equal to half intrinsic impedance of the free space. So the impedance matching to the 50 ohm characteristic impedance of the feeding line is difficult to achieve without using broadband baluns.

One of the possible designs uses simple bow-tie antennas with symmetric feeding lines, which is difficult to realize. Therefore, we propose a novel antenna which declines both drawbacks of a symmetric self-complementary antenna. The antenna is designed as an array of two halves of the bow-tie antennas above a ground plane as seen in the figure. The main advantage of the proposed design is its simple wideband matching and mechanical construction. Triangular monopoles are settled parallel one to each other at the distance of a half wavelength at the 3 GHz. The height of the monopoles is 50 mm, and the base is 100 mm. The in-phase excitation of the monopoles is performed by a microstrip line (2 mm wide and 1.6 mm high) which is fed in its center by a coaxial probe from the opposite side of the ground plane.

Antenna was designed and simulated by an electromagnetic solver, and then manufactured and measured. Experimental and theoretical results for the VSWR fits well. The proposed antenna shows good matching characteristics in the band 1 to 5 GHz with high radiation efficiency throughout the whole frequency band. Mutual impedance between monopoles does not degrade input impedance too match. Radiation pattern in the H-plane is almost omnidirectional at the lower end of the frequency band, and gradually changes to the bidirectional one at the upper end of the frequency band. Radiation pattern in the E-plane slightly changes with frequency.

Proposed antenna is suitable for UWB mobile communication applications.

Several fields could potentially benefit from the development of extremely broadband directive antennas. For example the ultra-wide and (UWB) frequency range 3.1-10.6 GHz, regulated by the Federal Communication Commission (FCC), demands for transmission and reception system capable to guarantee very little distortion of the signals (i.e. flat magnitude and linear phase of the transfer function represented by the transmitting/receiving antenna). Also the discipline of electromagnetic compatibility (EMC) presents formidable reasons to realize field sensors operating efficiently in broad and wide frequency ranges. The main motivation for this work was the development of a field sensor for electromagnetic compatibility to be used in the range 4-40 GHz. The antenna proposed for such sensor is a novel UWB directive and non dispersive lens antenna, the Pyramid Antenna.

This antenna is an extension of the wide band leaky wave antenna presented in [1]. It was based on the frequency independent radiation mechanism of the radiation occurring on a slot printed at the interface between two different dielectrics. The relevant Green's function has been theoretically investigated in [2] and [3]. If the two media are free space and a dense dielectric, radiation occurs primarily in the dielectric. This type of propagation does not suffer from the typical dispersion effects that characterize all previously proposed leaky wave radiation mechanisms.

A prototype of the Pyramid Antenna has been designed manufactured and measured showing impressive impedance and radiation performances. The input impedance is weakly frequency dependent. The reflection coefficient can be maintained lower than -10dB's over the bandwidth. The gain of the antenna is a design parameter: the minimum realistic gain is in the order of 8 dB's but the maximum is not limited. Once the gain is fixed the H-plane radiation pattern remains essentially frequency independent over a good part of the mentioned bandwidth. Thus also the overall gain is weakly varying with frequency. The phase center is well defined for observation aspects in the main beam and it is also frequency independent. CPW connected MMIC based T/R modules are easily integrated with the radiating slot.

[1] A. Neto, S. Bruni, G.Gerini, M. Sabbadini, "The Leaky Lens: a Broad Band, Fixed Beam Leaky Wave Antenna", IEEE Transactions on Antennas and Propagation, Vol. 53, no. 10, October 2005.

[2] A. Neto and S. Maci, "Green's Function of an Infinite Slot Printed Between Two Homogeneous Dielectrics. Part I: Magnetic Currents", IEEE Transactions on Antennas and Propagation, Vol. 51, no. 7, pp. 1572-1581 July 2003.

[3] S. Maci and A. Neto, "Green's function of an infinite slot line printed between two homogeneous dielectrics. Part II: Uniform asymptotic fields", IEEE Transactions on Antennas and Propagation, Vol. 52, no. 3, pp. 666-676, March 2004.

Knowledge of modes cutoff wavelengths in double ridged waveguides is highly demanded for determination of operation bandwidth for design of broadband antennas. It is shown that for nonstandard dimensions the TE₃₀ mode does not exist or is degenerated or the higher mode skips the TE₃₀ mode. A paper by Walton and Sundberg presents values of lambda_C¹⁰/a and lambda_C³⁰/a as a function of s/a for various values b/a and d/b. It can be seen that lambda_C³⁰ is relatively insensitive to changes in d/b, but lambda_C¹⁰ is strongly sensitive. The bandwidth increase rapidly as gap between the ridges becomes small.

For antenna design we should know not only cutoff frequencies of particular modes but also the field intensity distribution for TE and TM modes. CST Microwave Studio and HFSS were used for analysis, both results fit very well. The double ridged waveguide analysis in terms of b/a, s/a and d/b (the same as in the paper by Walton and Sundberg) was performed. It was found that it does not explicitly signify modes analogous to rectangular waveguide - the dominant TE₁₀ mode is regular from the point of view of an electrical field intensity distribution for all dimensions. Curves of lambda_C¹⁰/a are equivalent with curves by Walton and Sundberg.

On the other hand the study of higher modes shows mode degeneration or absence of the TE₃₀ mode for specific dimensions. Study was focused on modes that have a maximum of electrical intensity e.g. TE₃₀ mode between ridges. For specific dimensions of double ridged waveguide TE₃₀ mode not exists or is degenerated or higher TE_XX mode with maximum between ridges skips the TE₃₀ mode. We defined the maximum usable ridged waveguide bandwidth as the ratio of TE₁₀ to TE_X1 mode cutoff wavelengths. The TE_X1 mode is the first higher mode that has max. intensity between ridges.

The TE₃₀ does not exist when distance between a ridge and a side wall is equal to a half size of the side wall and at the same time a ridge width s is smaller than length of the side wall b. The curves of the TE_X1 mode were calculated and compared with by Walton and Sundberg. and the discrepancy was found. The possibility of the new attitude to the high quality (when compromise the bandwidth, against reflection, radiation pattern) of double ridged horn antennas design will be presented in a full paper.

This research is a part of the activities of the Department within the Antenna Centre of Excellence 2
WALTON, K.L., SUNDBERG, V.C., Broadband Ridged Horn Design, The Microwave Journal, March 1964, p. 96-101.

This paper deals with the time delay removal of an Archimedean spiral antenna which operates from 400 MHz to 4845 MHz. As it is known, spiral antennas show a dispersive behavior, thus, in time domain, a "chirp" pulse will be displayed. For time delay removal two procedures are investigated: the first one uses an error-term flow graph for the frequency signal as for Vector Network Analyzers (VNA) while the second one supposes to place a reference metallic plate at a certain distance in order to identify the phase dispersion given by the antenna. In the second case the received signal is passed in time domain by applying an ifft, the multiple reflection are removed and the phase variation due to the time propagation is subtracted. After phase correction the time domain response as well as the side lobes level is decreased. The antenna system made up of two Archimedean spirals is employed by a stepped frequency continuous wave radar which work with a frequency step of 35 MHz.

A spiral antenna is a dispersive system (fig 1). If a metallic plate is placed at a distance d of the antenna system and the power transmitted by the radar is equalized for all 128 frequencies then the transmitted signal can be written like:

where A is a constan.This signal propagates to the metallic plate and is scattered back to the antenna system. The received signal can be written as:where tin denotes the delay due to the antenna system and due to propagation towards the metallic plate and back; rn includes the propagation losses as well as the reflection losses.The delay tin is frequency dependent because the delays within the antenna system depend on the frequency. As the accuracy of the system depends on the synthesized pulse width the time delay within the antenna is a critical factor. The contribution of lower and high frequencies to increase the time response of the antenna system are analyzed in the paper. Then the VNA and single reference methods are described and compared against experimental results. The best results are achieved using a single reference calibration procedure which operates as a frequency domain matched filter (fig. 2). On figure 2 the distance is computed as dist=cti/2, where c is the speed of light and ti is synthesize pulse duration. The calibration procedure described in this paper has been successfully used for a stepped frequency radar used for landmine detection. The single reference method improved the time response of the antenna from around 9 ns to less than 1 ns, which drastically improved the down range resolution of the radar.

A wideband antenna that radiates a linearly polarized conical beam is investigated to realize a VSWR frequency band (VSWR < 2) ranging from 2.2 GHz to 15 GHz. The main element of this antenna (see Fig. 1) is a conducting body of revolution (BOR around the z-axis), which is backed by a circular conducting plane (a finite circular ground plane of diameter D_GP located in the x-y plane). The generating line of the BOR is expressed with an exponential function, which results in a BOR height and top diameter of z = z₁ and D_BOR = 2x₁, respectively. The BOR is used for radiating a linearly polarized conical beam.

A wideband VSWR characteristic is obtained by using a parasitic conducting ring, which coaxially surrounds the BOR. The ring has width W = (D_{RNG, out} - D_{RNG, in})/2, where D_{RNG, out} and D_{RNG, in} are the ring outer and inner diameters, respectively. The ring, located at z = z₁, is short-circuited to the ground plane with N conducting pins.

The antenna with a structure composed of the BOR, parasitic ring, and pins is abbreviated as the BOR-PaRP antenna. The configuration parameters of the BOR-PaRP antenna are D_GP, 2x₁, z₁, D_{RNG, out}, D_{RNG, in}, and N, as mentioned above. Among these, only the BOR top diameter 2x₁ and the ring inner diameter D_{RNG, in} are varied subject to the objectives of the analysis. The other parameters (D_GP, z₁, D_{RNG, out}, N) are fixed throughout this paper.

Effects of the parasitic ring width W on the VSWR are investigated by varying the ring inner diameter D_{RNG, in}, while keeping the BOR top diameter 2x₁ constant. In this case, as D_{RNG, in} is decreased, the spacing between the BOR and ring, that is, (D_{RNG, in} - 2x₁)/2 = s_BOR-RNG, is also decreased. The investigation is repeated with numerous 2x₁. The analysis reveals that the decreased spacing improves the VSWR at low frequencies. Fig. 2 shows the VSWR for optimized 2x₁ and D_{RNG, in}, where a wide VSWR frequency band ranging from approximately 2.2 GHz to 15 GHz is obtained. Within this frequency band, the antenna radiates a linearly polarized conical beam. Note that the BOR height and ring outer diameter are z₁ = 1 cm and D_{RNG, out} = 4 cm, respectively. The detailed configuration parameter study is presented in the paper.

The present paper describes a cooperation between a number of European partners under the European Network of Excellence, ACE. The purpose of the cooperation is 1) to identify the state of the art within Europe concerning reflector surface model-ling, 2) to reveal the topics where the synergy between several partners can provide results which were otherwise impossible and 3) to detect possible areas where further work is needed.

Reflector antennas find use in terrestrial radio links and as an-tennas for satellite communications – both on ground as termi-nals and onboard the satellite. A standard design of a reflector antenna normally assumes that the surface is perfectly conduct-ing. However, modern reflector antennas are often equipped with special surfaces to obtain a particular performance such as polari-sation control using a gridded reflector or frequency diplexing us-ing a frequency selective surface. Special treatment is also neces-sary to account for special materials such as CFRP, tricot mesh for unfurlable reflectors or paint. Also special surfaces are used as radomes to protect antennas under adverse climatic or environ-mental conditions, on vehicles, trains and aircraft, or onboard satellites.

For the design of a reflector antenna it is essential that fast and reliable software is available to model the local reflection and transmission properties everywhere on the surface. The paper will illustrate the validity of approximate methods for strip grids and rectangular meshes by comparing them to accurate Method of Moments solutions.

The global radio astronomy community has for the last several years embarked on an exciting project known as Square Kilometer Array (SKA). This project is a radio telescope operational from 0.1GHz to 25GHz, with a square kilometer of overall collecting area, providing an increase in the sensitivity of two orders of magnitude. To meet this wideband requirement, many concepts from several national institutes, have proposed various concepts which will be reviewed in this paper. Until recently many different concepts have been studied, which included the Aperture Array (phased array) concept as well as the reflector based dishes of various sizes, including the Chinese 500 m diameter dish. While some of these concepts are still being studied as potential candidate antennas, the SKA project now emphasises concepts with large number of smaller antennas. These conceptual ideas are now incorporated in the Reference Design (RD) document produced by the International SKA project Office (ISPO). The RD suggests that the reflector based solutions may be appropriate above 0.5 GHz whereas the Aperture Array solution will be useful below 1.5GHz.

Almost all the proposed concepts use the traditional reflector approach with either a very wideband feed, or a cluster of combined feed to form a focal plane array, which are also known as phased array feeds. The focal plane array concept has only recently been proposed and at present, appears to show much promise of broadband capability with multiple fields of view (FOV).

One concept which is quite different from the above is the Aperture Array (phased array) concept. This concept although well established in radar and communications, is novel for radio astronomy. The architecture of the aperture array is configured such that it allows us to provide simultaneous multiple fields of view from a single aperture. This concept although originally proposed by ASTRON, has been promoted by the European radio astronomy community, as the European concept for the SKA. The European Commission (EC) is funding the development of the Aperture Array concept through the Sixth Framework (FP6) project known as SKA Design Studies (SKADS), which is co-ordinated by ASTRON. Although SKADS project contains many additional scientific studies, the major component of SKADS is the designing, testing of the Phased Array. The Aperture Array would be operational below the 1.5GHz region, where some of the Key Science Projects (KSP) can be carried out.

Within the SKA community, the proposers of the various concepts are now required to provide a proof of concept by providing the results of design, development and test. ASTRON is at present involved in such a task with their European colleagues.

The key element of radio telescope is the very wide bandwidth antennas and ASTRON has been actively working in this area for several years. Astron has demonstrated various capabilities on a smaller scale culminating in the present project of building a large demonstrator of 300 m2. This paper will also review and provide a summary of antenna development at ASTRON leading to the present EMBRACE project using Vival

Arrays of Vivaldi tapered slot antennas have become an established technology for wide bandwidth scanning arrays. Antenna bandwidths of 10:1 with scanning to 45 deg are achievable. To design the array, mutual coupling must be properly included and, in fact, utilized. Therefore, problems have arisen when these arrays are fabricated as independent subarrays without electrical contact between adjacent elements or subarrays. Nevertheless, it is highly beneficial to consider arrays with gaps because it is much easier to fabricate and to repair the array if the elements can be inserted or removed individually or in small groups. This paper presents results of some recent studies of Vivaldi antenna arrays wherein there are gaps between elements and/or subarrays. The gaps are assumed to be sufficiently large that element-to-element coupling across the gaps is much different than coupling between elements that are electrically connected to their neighbors. Such gaps between all elements of an array usually produce severe impedance anomalies that disrupt the operating band of the array. Employing mirror symmetries in arrays of Balanced Antipodal Vivaldi Antennas has eliminated these anomalies over a substantial portion of the operating frequency range with no apparent detrimental effects. Gaps between subarrays of various sizes also have been studied. Plots showing the effects of gap width and subarray size will be presented to indicate the importance of these parameters to array fabrication and performance.

Broad-band wide-scan planar phased arrays are suitable for multiple applications and are useful both in military and commercial sectors. For airborne applications the size of the antenna is important, and benefits with thin and light-weight antennas are evident. The size of the antenna aperture of the phased array is restricted by the desired directivity, so to further reduce weight and cost multi-role capabilities are required. A multi-role phased array antenna needs to be capable of broad-band and/or wide-scan capabilities to fit tasks such as radar, data links and electronic warfare.

An antenna element that is promising for multi-role phased array antennas is the fragmented aperture element. This antenna element consists of a conducting pattern etched on a dielectric backed by a groundplane. The conducting pattern, dielectric thickness and permittivity are designed with the help of a genetic algorithm (GA). The cost function is evaluated using a finite-difference time-domain (FDTD) code with periodic boundary conditions [1]. Fragmented aperture arrays have been studied earlier [2] with focus on broad-band properties of the array out to a scan angle of 45° for |Ã|≤-10 dB (E-plane and H-plane). In [2] each fragmented pixel was represented by a single FDTD-cell. It was shown that a coarse grid for the FDTD-cells agreed well with higher resolution meshes for broadside scan. However, when the array was scanned in the E-plane and H-plane the performance changed for the worse and reflection coefficient increased [3]. Hence, the requirement was no longer possible to meet.

In this paper the focus is to study the wide-angle performance of these array elements. This requires a higher resolution of the FDTD-grid than previously. Numerical results on these antenna elements will be presented for larger scan angles than 45°. One example is found in Figure 1 where substrate thickness was fixed and pattern, dielectric constants and superstrate thickness was optimized.

There are also some practical problems that need to be solved before manufacturing the array element. Earlier the dielectric constant of the substrate was a parameter that could change linearly; this resulted in materials that were not commercially available. The simulations will now be restricted to a list of commercially available dielectrics. Furthermore, another problem was that the corners of diagonally adjacent pixels in the aperture are electrically connected and may be difficult to manufacture. To avoid manufacturing problems two strategies are investigated to avoid connection problems. The motivation for this is to make necessary computations before manufacturing a fragment patch array, which will be done in the near future.

[1] H. Holter and H. Steyskal, "Infinite phased-array analysis using FDTD periodic boundary conditions-pulse scanning in oblique directions"IEEE Antennas and Propag., vol. 47, no. 10, pp. 1508-1514, Oct. 1999.
[2] B. Thors, H. Steyskal and H. Holter, "Broad-band fragmented aperture phased array element design using genetic algorithms"IEEE Antennas and Propag., vol. 53, no. 10, pp. 3280-3287, Oct. 2005.
[3] H. Steyskal and H. Holter, Private communication,2006.