Multi spot beam scenarios with a so called "four-color" topology for the coverage area are currently one of the most popular concepts for modern telecommunication satellites in Ka-band. Usually they are realised as single feed per beam (SFB) architectures. Major disadvantage of this concept is the need of eight relatively large reflector antennas in order to generate the adjacent beam topology [1][2]. Using combined Tx/Rx feed chains the number of reflectors can be reduced to four. However, because of the wide spacing between the Ka-Tx-band and the Ka-Rx-band the design of the feed chain is not easy and the diameter of the reflector is a difficult compromise between the optimal sizes for Tx and Rx. Another solution is the use of oversized shaped reflectors [4]. Only one reflector for Tx and one for Rx is required but the reflector is that big, that it can not be accommodated as a solid shell reflector on current space crafts.
An alternative to the SFB concept are multiple feeds per beam (MFB) antennas. This means for each beam a number of feeds is used. Each beam has now a feed cluster instead of a single feed. A feed horn may be used for different clusters. In this way an overlapping of adjacent clusters can be realised. Two basic principles exist for MFB antennas, space fed arrays (waveguide lenses) and arrays fed by a beam forming network [5].

In this paper we investigate a MFB antenna fed by a BFN. A typical Pan-European coverage with more than 80 beams was chosen as scenario. For this coverage a Tx and a Rx antenna was optimised for directivity, cross polarisation and isolation. Contour plots of the achieved performance are presented. Assessments are made for the electrical performance and the mass of the feed network. All data show that MFB antennas, with only one reflector for Tx and one for Rx, can achieve similar or even better electrical performance than SFB antennas using four reflectors. Based on these promising results it was decided to manufacture a demonstrator model to verify the predicted performance by measurements.

[1] Rao: "Design and Analysis of Multiple Beam Reflector Antennas", IEEE Antennas and Propagation Magazine, Vol. 41, No. 4, pp. 53-59, August 1999
[2] Rao: "Parametric Design and Analysis of Multiple-Beam Reflector Antennas for Satellite Communications", IEEE Antennas and Propagation Magazine, Vol. 45, No. 4, pp. 26-34, August 2003
[3] Rao s., Chan, K.K., Tang, M.: "Dual-Band Multiple Beam Antenna System for Satellite Communications", Antennas and Propagation Society International Symposium 2005, Vol. 3A, pp. 359-362
[4] Balling, P., Mangenot, C., Roederer, A.G.: Shaped Single-Feed-per-Beam Multibeam Reflector Antenna, Proceedings EuCAP 2006
[5] Gehring, R. et al: Trade-Off for Overlapping Feed Array Configurations, 29th ESA Antenna Workshop, 18 - 20 April 2007, ESTEC, Noordwijk, The Netherlands

Multiple Beam antennas are particularly interesting in satellite telecommunication systems in order to increase the capacity by frequency and polarisation reuse over the full coverage. Typical passive accommodation for a reuse scheme of 4 requires 4 reflectors. To improve the antenna performances, one of the authors suggested to implement beam-hopping in a standard reuse scheme of 4 in order to produce a reuse scheme equivalent of 12 or even 16 [1]. The idea is to reduce the effectively used area covered at a given instant ; thus without changing the size of the reflector, one can virtually increase the minimum directivity of the antenna, and as a consequence the C/I should also increase. The interest of the concept has been investigated in [2]. When compared to the typical accommodation antennas performances, the proposed technique enables to increase up to 3 dB the minimum directivity and C/I.

Previous investigations were performed on simplified models that did not consider the constraints linked to the beam hopping technique implementation [2]. In this paper, we will describe a more realistic FAFR antenna, where each sub-spot is produced by a sub-array of 7 feeds. Switching between the sub-arrays produces the beam-hopping. Good performances were found for what we call the higher reuse factor techniques, particularly in the case of a reuse scheme equivalent to 12 (k = 12 case). But for the sectorization technique, the geometry of the focal array is not fully compatible with the desired spots and performances are under expectation. For the k = 12 case, the C/I is improved by more than 4 dB, while the minimum directivity is increased of 1 dB.

[1] J. Sombrin, "A Roadmap for Commercial Flexible Ka-Band Telecommunication Satellites", 11th Ka and Broadband Conference, September 25-28, 2005, Rome, Italy.

[2] N. J. G. Fonseca and J. Sombrin, "Multiple Beam Antennas Using Beam-Hopping and Size Reduction of Effectively Used Spots for Improved Performances", 29th ESA Workshop on Multiple Beams and Reconfigurable Antennas – Innovation and Challenges Conference Proc., April 18-20, 2007, Noordwijk, Netherlands, pp. 348-351.

The possibility of employing the same spacecraft at several orbital locations, changing the coverage region, and compensating for varying weather conditions are the main reasons why reconfigurable antennas for contoured beam generation can provide significant advantages in satellite communications. Though extremely attractive in terms of cost and manufacturing, shaped reflectors lack so far the capability of being reconfigured in orbit, while this feature can be obtained by an array-fed parabolic reflector.

Several examples of possible reconfigurable reflectors were studied over the years. Today the most promising technology is constituted by a mesh of interwoven flexible wires with circular cross section which is supported by a number of control points and has a free rim.

While the mathematical properties and a preliminary physical realization of such a model were described in previous articles, we investigate here its use in a realistic mission scenario in Ku band. A dual reflector typical for space applications will be considered. Repointing of the antenna and mechanical reconfiguration of the reflectors for the different coverages will be assumed. In order to evaluate the advantages and limitations provided by the use of a reconfigurable reflector and to define preliminary guidelines for the number and position of the actuators, the performances given by a traditional fixed shaped reflector designed to illuminate the envelope of the desired coverages will be first analyzed. This represents today the best obtainable solution when reconfigurability of the antenna can not be achieved. Second, the results given with a fixed main reflector and a fixed subreflector, both shaped for each coverage, will be shown, providing the best performances when complete reconfigurability is possible. Finally the performances given by a shaped but fixed main reflector and reconfigurable subreflectors, one for each coverage, will be presented.

Present and next generation telecommunication satellite systems often require multiple beam capability. A simple way to achieve high edge of coverage gain is to use a single aperture and a single feed per beam [1]. In the series of works [2], [3], [4] an approach based on the use of dielectric super-layers to enhance the radiation properties of small apertures to improve the excitation of reflector antenna systems has been proposed. In particular [4] shows the results for a demonstrator used as feed of a dual reflector characterized by moderate values (0.85) of the F/D (Focal distance to diameter ratios). In this contribution the same strategy based on the use of super-layers is adapted to the design of reflector systems based on larger F/D's (1.7) which in turn are necessary to achieve small degradation of the performances for the off focus beams. The inter-element distance now considered is large (2.4 wavelengths) which is typical for satellite based multi beam telecommunication applications. The super-layer can be realized with single dense dielectric slab (εr = 20.25) a double dielectric slab (εr = 4.5). The calculated embedded patterns provide an increase of the edge of coverage gain, with respect to the free space case, of at least 2. dB in an operating bandwidth of 1.7%. During the oral presentation the hardware manufactured to implement the previous concepts, including the filters for the frequency and reuse scheme, will be shown. Also the measured results will be compared with the simulations.

References
[1] S. K. Rao, "Design and Analysis of Multiple-Beam Reflector¡", IEEE Antennas and Propagation Magazine, Vol. 41, Aug. 1999, pp. 53-59.
[2] A. Neto, N. Llombart, G.Gerini, M. Bonnedal, P. De Maagt ,"EBG Enhanced Feeds for the Improvement of the Aperture Efficiency of Reflector Antennas", IEEE Transactions on Antennas and Propagation, Vol. 55, no.7, August 2007
[3] N. Llombart, A. Neto, G.Gerini, M. Bonnedal, P. De Maagt, "Impact of Mutual Coupling in Leaky Wave Enhanced Imaging Arrays", accepted for publication in the IEEE Transactions on Antennas and Propagation
[4] N. Llombart, A. Neto, G.Gerini, M. Bonnedal, P. De Maagt, "Leaky Wave Enhanced Feed Arrays for the Improvements of the Edge of Coverage Gain in Multi Beam Reflector Antennas", accepted for publication in the IEEE Transactions on Antennas and Propagation

Multibeam reflector antenna using the one feed per beam concept with only one aperture suffers from the well known problems of spillover losses or beam crossover losses. Some studies have demonstrated that a multifeed EBG resonator antenna can achieve passively on its surface directive and interlaced focal feeds. Thanks to this particular property, only one reflector antenna feeding by an EBG structure is needed to realize an efficient multispot coverage.

This paper deals with the design of an EBG antenna feeding a Side Front Offset Cassegrain Antenna (SFOCA) to generate a high gain European multibeam coverage in Ka band for up link multimedia telecommunications. The EBG antenna must be directive (24 dB) on 4% bandwidth and must present very low sidelobes (-30 dB). In order to obtain these difficult specifications the EBG antenna has been improved thanks to an efficient feed which is conical horn. This last permits to double the radiation bandwidth and to decrease the side lobes (-15 dB). The second job has been to avoid the parasitic effects caused by the no excited accesses. Consequently, a new EBG antenna with two levels has been designed in order to deport the disturbing accesses outside the EBG cavity. This concept permits to have the same results in multifeed configuration (directivity 24 dB and sidelobes - 30 dB) than the one feed configuration. As the performances of the SFOCA fed by this EBG antenna agrees with the specifications (GEOC and C/I), an EBG antenna prototype has been realized.

Antenna systems based on reflectors have a broad field of applications over the telecommunications satellite industry. Several space applications have been developed in few past years based in this technology. In the same way, this subject has been object of innumerable studies and research activity.

A typical set up for a feed cluster is made, for instance, with Horn Antennas, these radiating elements have demonstrated a high performance for a huge number of applications. Nevertheless, in recent applications like multi-beam based systems, Horn based feed clusters solutions present limitations in terms of compactness (i.e. number of required reflectors). These drawbacks, in the majority of cases, are due to the required radiating area of the horn antennas to achieve efficient spill-over performances that made it impossible to space out the radiating elements sufficiently close to obtain the desired beam lattice. One possible solution could be to change the type of primary feeds used to illuminate the reflector, so for instance, an array of small radiating elements could have the same radiation properties of a larger Horn antenna and the single-feed-per-beam feed cluster could be replaced by an array of radiating elements with each element participating to the formation of different beams. However, a system of these characteristics, termed Array Fed Reflector, brings with it several challenges related on how to feed the array without increasing the complexity of the system, and at same time, fulfil all the requirements of a set of single-feed-per-beam antennas (typically 3/4) in a single reflector architecture.

The system that is introduced here, named 2-D Double FFT-BFN, is a two-dimensional Beam Forming Network that uses the properties of the Fourier Transform to reduce its complexity, given the possibility of replace a cluster of antennas, used for instance, in Multi-beam reflector-antenna systems, by an array of electrically-small antennas with similar radiation characteristics. Additionally, a 2-D Double FFT-BFN could make possible the overlapping of the beams at aperture plane, thus, improving the focusing characteristics of a normal reflector system.

The BFN based on the application of twice the Fourier Transform over a unique input signal are designed to perform a physical movement of the spot beam defined at aperture plane. The main idea behind this system is to take advantage of the Fourier Transform Properties to reduce the electronic and complexity associated to a traditional BFN with similar characteristics.

The system proposed consists on two backed two-dimensional FFT circuits interconnected between them by a reduced group of “useful” signals. The system behaviour could be better understood from its 1-D representation. A complete system is composed by a set of feeding points that fed a FFT module of dimension MxM, from the M outputs obtained from this block we select only k elements, the rest of elements are connected to matched loads. The set of the named useful signals are now used as feeding for the second FFT block, this time of dimension NxN. After the second FFT block we obtain the desired amplitude and phase necessary to feed a sub-array of selected radiating elements.

Any desired shape or correction of the radiation parameters can be easily achieved applying small changes to the set of signals defined as interconnection signal. For instance, defining a Gaussian shape at this state shall result in a Gaussian shape on the illuminated area. This means that the radiation beam could present less side lobe level as obtained with a Flat-Top amplitude.

Figure 1. Schematic representation of a 1-D Double FFT-BFN

Figure 2. Schematic representation of a 2-D Double FFT-BFN

The paper will report the activities ongoing at ESA about the development of an advanced payload/system software simulation tool for evaluating at system level the overall performance degradations due to linear and non linear distortions generated in multibeam satellite communication payloads. As an example of application, a simulation of a multibeam payload based on advanced Tx active reflector antenna with a novel redundantless high power output section using "flexible" TWTAs is proposed. The results of these simulations will provide useful information about the potentialities of this advanced flexible output section/antenna in terms of overall distortions, optimum TWTA Output Back-Off and associated on board power efficiency.

In this paper an alternative multibeam antenna system, based on an active lens is presented. It may allow improving the performances of conventional multibeam antennas based on reflectors.

The future generation of communication satellites will use multiple beam antennas (MBAs) providing wide band two ways communication applications. High gain multiple overlapping spot beams, adopting both frequency and polarisation reuse, will provide the needed coverage. In order to generate several high gain spot beams, electrically large antenna apertures are required. These apertures are today mainly realised by reflectors.
Most of the operational or planned multiple beam antennas adopt, for up and down links, one feed per beam architecture with adjacent beams generated by different reflectors fed by a cluster of horns. This leads to three to four reflector antennas for European or CONUS coverage receive functions and the same number for transmit [1]. This multi aperture antenna architecture is the one today usually considered since, when a single aperture is used, inter-feed spacing requirements and feed diameter lead to inefficient illumination of the reflector and insufficient performance in terms of isolation levels. Each feed being associated to a beam, beam-forming networks are avoided. However, the volume required for the set of reflectors is very large, the system does not provide for any flexibility in terms of coverage and the reflectors are difficult to accommodate on the spacecraft.
A possible solution to generate a multi-beam coverage using a single aperture is the Focal Array Fed Reflector described in [2]. This concept is based on overlapped beam footprint in the reflector focal plane. This overlapping is performed by connecting together individual feeds using a beam-forming network, some of these feeds being used for several beams. This antenna concept is quite complex at focal array level but, with limitation on the maximum number of beams, has the advantage to allow the generation of different beams sizes and shapes with only one aperture.
An alternative approach, based on a single aperture, consists in overlapping contiguous feeds in a completely radiative way [3-4].
All these concepts are suitable candidates for multi-beam applications but suffer from different implementation difficulties. In a long term perspective, simpler alternative solutions based on a single aperture are needed.

An antenna system based on a discrete lens could be ideally used for multi-beam communication payloads in reception or in transmission. Discrete lenses are more tolerant to dimensional inaccuracies and can be mounted axially, providing a high tolerance to torsional and warping distortions (reflectors are usually mounted in offset configuration). Besides, Discrete Lens Antennas (DLAs) can be designed for low side-lobe levels at large steering angles. Beam pointing is implemented without phase shifters. As a consequence a low complexity antenna which guarantees an easy deployment and satellite accommodation may be obtained. Active elements may be included inside the lens in order to obtain a fine in-orbit tuning capability for deployment, pointing, manufacturing inaccuracy and ageing-caused performance degradation with high reliability due to graceful degradation. DLAs allow low sensitivity to multipaction thanks to power distribution in a radiative way.
DLA are composed by 2 arrays of N elements each, one facing the radiating side where the beams are generated in transmission or reception, one facing the feed side i.e. in front of the M feeds placed on the focal area. As evident in Fig.1, every element in the radiating side is connected with a corresponding element in the feed side via a variable length transmission line. The length of the delay lines varies across the array in such a way that an incident plane wave is focused in a focal point in the near field in the feed side.
An interesting DLA architecture is the one proposed by McGrath [5] based on two parallel and planar arrays and shown in Fig.2. The particular geometry reported in this picture has been defined to achieve a beam width of 0.6°. By using 3 square apertures for the array of the radiating side about 1500 radiating elements are required for each array. This planar DLA can be fabricated as a thin deployable panel structure using standard circuit technologies. The resultant beams have high directivity and good isolation. The main drawback of this configuration [6] is due to the missing alignment between the radiating elements of the array facing the feed side and the feeds on the lens focal zone. To avoid this problem a large number of very small radiating elements (0.8÷1 ,  being the wavelength), with dual polarisation and low cross-polarisation, arranged in sub-arrays having different beam pointing could be used, yielding the resulting DLA configuration too complex.

REFERENCES [1] S.K. Rao, “Parametric design and Analysis of multiple beam reflector antennas for satellite communications”, IEEE Antennas and Propagation Magazine, Vol. 45, No. 4 August 2003, pp. 26-33. [2] C. Mangenot, P. Lepeltier, J.L. Cazaux, J. Maurel, “Ka-band fed-array focal reflector receive antenna design and development using MEMS switches”, JINA Conference 2002, Vol. II, 12-14 Nov. 2002, pp.337-345 [3] R. Chantalat, P. Dumon, B. Jecko, M. Thevenot, T. Monediere, “Interlaced feeds design for a multibeam reflector antenna using a 1-D dielectric PBG resonator”, IEEE AP-S International Symposium and USNC/CNC/URSI National Radio Science Meeting, June 22-27, 2003, The Ohio State University Columbus, Ohio. [4] S. I. Khureim Castiglioni, G. Toso, C. Mangenot, “Multi-beam antenna based on a single aperture using overlapped feeds”, JINA 2004 Conference, Nice, November 2004. [5] D.T. McGrath, “Planar Three-Dimensional constrained lenses”, IEEE Transactions on Antennas and Propagation, Vol. 34, No. 1 January 1986, pp. 46-50. [6] G. Ruggerini, “Antenna multifascio a lente discreta e planare per satellite di telecomunicazioni”, Progettazione e Sviluppo di Moderni Sistemi di Antenna, XIII Giornata MECSA, Università di Salerno, Fisciano 14-15 Maggio 2007,pp 29-32. [7] A.K. Bhattacharyya, G.Goyette, “A Novel Horn Radiator With High Aperture Efficiency and Low Cross-Polarization and Applications in Arrays and Multibeam Reflector Antennas”, IEEE Transactions on Antennas and Propagation, Vol. 52, No. 11 November 2004, pp. 2850-2859.