Charles W. Bostian
Vincent W.S. Chan
E. Paul Hager
Neil R. Helm
Christoph E. Mahle
Edward F. Miller
Joseph N. Pelton
"Investing in antenna technology always pays. This is a very crucial technology." Giorgio Perrotta, Alenia Spazio.
An antenna is a "transducer" between electromagnetic waves in space and voltages or currents in a transmission line. When transmitting, the antenna converts electrical signals into radio waves; a receiving antenna reverses the process and transforms radio waves back into electrical signals. Most antennas are passive, simply metal structures that launch or collect radio waves.
All passive antennas can transmit and receive, and a principle called "reciprocity" allows a passive antenna's transmitting properties to be derived from its receiving characteristics and vice versa. All antennas are directional, transmitting more power in some directions than in others. The direction of maximum transmission or reception is called the "boresight direction." The directional properties of an antenna are described by its radiation pattern, a pictorial representation of relative radiated power versus direction. Radiation patterns are three-dimensional. In practice, radiation patterns are usually drawn to show relative radiated power in two orthogonal planes, a horizontal plane containing the antenna (the azimuth plane) and a vertical plane passing through the antenna (the elevation plane). Figure 2.1 is an example.
Radiation patterns exhibit a lobe structure in which the largest lobe in three-dimensional space constitutes the beam through which the antenna radiates and/or receives most of its power. In this context, the locus of half-power points surrounding its peak is taken to be the perimeter of the main beam. The intersection of the main beam with the ground constitutes the "footprint" of an antenna on a satellite.
The tendency of the antenna to concentrate its radiated power is called "gain." The angular width of the main beam measured between the half-power points is called the half power beamwidth (HPBW) or more colloquially, beamwidth of the antenna. Beamwidth and gain vary inversely with each other; a high-gain antenna has a narrow beamwidth and vice versa. Gain is proportional to the physical area of an antenna expressed in square wavelengths. The size of an antenna is often stated in terms of its "aperture," which can be taken to mean either the physical area of the radiating part of the antenna or the diameter of a circle having this area, depending on context.
Radiation pattern lobes other than the main beam are called "sidelobes." The higher the sidelobe level, the more likely an antenna is to interfere with or be interfered with by a receiver in the direction of the largest sidelobe. For this reason, the sidelobe envelopes of ground antennas used in the U.S. must meet specifications set by the Federal Communications Commission (FCC) to insure that an antenna pointed at one satellite does not have an unacceptably high sidelobe pointed at another satellite separated from the first satellite by 2 degrees or more in the geostationary arc.
Electromagnetic waves are vector quantities, that is they are polarized. The preferred (desired) polarization radiated by an antenna is termed co-polarization; the orthogonal polarization is termed cross-polarization. If polarization can be made pure enough, orthogonally polarized waves can travel together without interference and be separated by properly designed receiving antennas. The ability of antennas to discriminate in this manner is measured by the cross-pol ratio, or polarization isolation.
An antenna has a "center frequency" or "operating frequency" -- the frequency at which the antenna exhibits optimum performance -- and a bandwidth -- the frequency range over which the antenna impedance and radiation pattern remain within some required tolerance.
A good antenna design is one that achieves the required gain, sidelobe level, polarization isolation and bandwidth specifications in the smallest aperture size consistent with mass limitations and ease of manufacture, all at an acceptable price.
Figure 2.1. A Radiation Pattern (Courtesy MBB Deutsche Aerospace)
Antenna technology is a mature field. In recent years progress has been evolutionary, driven by the availability of (a) new software tools for analyzing and designing antennas, (b) new materials like high temperature superconductors, carbon fibers, and advanced substrates, (c) new components like GaAs monolithic microwave integrated circuits (MMICs) and opto-electronic devices, (d) new measurement methods, and (e) new manufacturing techniques, as well as (f) new needs and priorities. All these factors influence antenna designers; the extent of their influence depends on whether the resulting increases in system capability justify the costs.
Most of the mathematical theory of antennas has been available for many years, but the impossibility of solving the equations by classical, exact, closed-form methods made antenna design an empirical art. Advances in computer hardware and software over the last 25 years have gradually changed this, making it easier to design most standard antennas (reflectors in particular) and automating the process of verifying the performance of prototypes.
Recent commercial space antenna research has focused on (a) techniques for developing accurately shaped radiation patterns, so that a satellite's footprint can be tailored to fit a desired geographic region, (b) techniques for increasing frequency reuse through orthogonal polarization isolation and beam to sidelobe isolation in multiple beam configurations, (c) ways for modifying the radiation pattern in orbit, so that the footprint can be adjusted for changes in satellite location or traffic demand, and (d) developing the particular radiation patterns (multiple spot beams which can be scanned rapidly) that mobile systems and some inter-satellite links (ISLs) require.
Reflector Antennas. Traditional geostationary (GEO) satellite fixed satellite service (FSS) and maritime mobile satellite service (MMSS) links at L-band, C-band and Ku-band require antennas with high gain. Parabolic "dishes" have been the design of choice. These consist of a paraboloidal reflector illuminated directly by a set of "feed" antennas or indirectly through a system of subreflectors. The directly illuminated version is called a "prime focus fed" antenna. Indirectly illuminated versions are usually based on classical "folded" optical telescopes -- "Cassegrain" (or "Cassegrainian") and "Gregorian" antennas for example. See Figure 2.2.
Feed antennas are usually horns (Figure 2.3). In general, single horns are used to develop relatively simple circularly symmetric beams, while multiple horns (Figure 2.4) are used to create "shaped" beams. These multiple horns constitute a phased array (see below), although not usually described as such.
Phased Arrays. An alternative approach to large reflectors is to deploy an array of small antennas over a large area and connect them in such a way that their received or transmitted signals are in the correct electrical relationship (phase) with each other, i.e., a "phased array" or simply an array. The individual small antennas are called "elements." Small arrays, called "subarrays," can be combined to form larger ones. Figure 2.5 shows a phased array.
The circuitry which does the phasing in a phased array is called a "beam forming network" (BFN) (Figure 2.4). In "passive" arrays, the BFN is an assembly of transmission lines, phase shifters, attenuators, and other "plumbing." If power amplifiers, low noise amplifiers, or other electronic devices are included as an integral part of the BFN, the antenna is called an "active" array.
Figure 2.2. 18.3 m INTELSAT Standard A Earth Station (Courtesy ANT Bosch Telecom)
Figure 2.3. A Feed Horn and Associated Network
Figure 2.4. A Multiple-Horn Feed System and the Accompanying Beam Forming Network (Courtesy Alenia Spazio)
Figure 2.5. A Phased Array Antenna (Courtesy D.M. Pozar and D.R. Jackson)
Scanning and Reconfiguration. Moving the beam of an antenna in a particular direction is called "scanning." Reflector antennas are usually scanned by mechanically moving the reflector or the feed or both. A phased array can be electrically scanned by changing the phasing of its elements.
Changing the shape of an antenna beam or changing the number of beams is called "reconfiguration." Since this process can also change the direction of the beam or beams, there is no sharp distinction between scanning and reconfiguration, and the terms are sometimes used interchangeably. However, reconfiguration usually applies to an occasional process, whereas scanning usually is performed on a more frequent, often regular basis. On-orbit reconfiguration is often accomplished by having multiple layer BFNs, one for each desired beam configuration, and switching to the desired configuration on ground command. For example, INTELSAT uses three-layer BFNs to achieve three-region coverage with the same feed-horn-reflector configuration.
Array-fed Reflectors. Many variations on the basic reflector and phased array themes are possible. Phased arrays can radiate directly into (or receive from) their environment, i.e., "direct radiating arrays." Phased arrays can also be used to illuminate reflectors. This combination can be electronically scanned or reconfigured.
Array feeds can compensate for the inherent distortions of large reflectors deployed in space, both by techniques that monitor the reflector surface or by adjusting the feed to synthesize a desired radiation pattern (Rahmat-Samli 1990).
Most current low earth orbit (LEO) satellites lack space to deploy large antennas and instead carry simple monopole antennas, often called "whips." The reduction in antenna gain in going from the reflector antenna carried by a typical GEO satellite to a whip is compensated for by the reduction in path loss resulting from shrinking the radio path length from perhaps 40,000 km for a GEO satellite to perhaps 2,000 km for a LEO satellite.
Ground antennas for LEO systems have tended to be yagis or helices, because, at the low frequencies usually used by LEO satellites, dish antennas would be impractically large. But there is no inherent difference in the requirements for LEO and GEO system ground antennas, and, with the advent of LEO systems like Motorola's IRIDIUM that require sophisticated beams, LEO satellites may soon be carrying phased arrays and reflector antennas.
Smaller Earth Station Antennas. Early satellite links used very large ground antennas to compensate for low-powered transmitters, noisy receivers, and the relatively small antennas that could be carried on satellites. Thus 30 m (diameter) antennas became standard in early INTELSAT earth stations, and the earth station antenna was a major cost factor in any satellite communications system.
Since the cost of a parabolic reflector antenna varies in proportion to something between the diameter squared and the diameter cubed, significant savings can be and have been achieved by reducing the size of the initial, very large terrestrial antennas. Quieter and more sensitive receivers, bandwidth reduction, higher power transmitters and higher gain antennas in space, and increasingly sophisticated modulation and coding schemes all led to the use of smaller antennas on the ground (as well as the increasing use of Ku-band, where larger values of downlink radiated power density are allowed than at C-band), to the point that the antenna of a modern VSAT (very small aperture terminal) can be less than one meter in diameter and is not a major factor in the terminal's cost (see Figure 2.6).
Figure 2.6. A Prototype Suitcase VSAT Terminal Developed for the European Space Agency (ESA) by Space Engineering and Schrack Aerospace (Courtesy ESA)
The major principle behind this size reduction in ground antennas is that, in the logarithmic "power budget" equation that defines a communication link's performance, the transmitting and receiving antenna gains and the transmitter power are all additive, and the system noise temperature is subtractive. Within limits, these quantities can all be traded off against each other. So long as their sum is constant, system performance is unchanged. A practical limit on the gain of a satellite antenna is imposed by the need to keep the beamwidth large enough to cover the desired geographical area. A further principle is that sophisticated modulation and coding schemes lead to coding gains which also add to the system gain or permit smaller antenna diameters.
Horn-fed reflector antennas constitute a mature technology which is particularly suited to those applications where an earth station or a satellite communicates with space or ground terminals in relatively fixed locations where there is room to install a dish. Thus most satellites carry front-fed offset paraboloid antennas (Figure 2.7). Spot beams are generated by exciting appropriate collections of feed horns. Two reflectors may be nested, one behind the other, to provide beams with different shapes on, say, vertical and horizontal polarization. The front reflector would reflect horizontally polarized signals while being nearly transparent to the orthogonal vertically polarized signals. The rear reflector would reflect the vertically polarized waves that passed through the front reflector.
Figure 2.7. A Front-Fed Offset Reflector Antenna with Multiple-Feed Horns (Courtesy Alenia Spazio)
Reflector antennas for earth stations have reached a similarly mature state of development. Both prime focus and Cassegrain models are widely available at reasonable costs. It is hard to imagine realistic developments in ground antenna technologies that would significantly change the competitive situation so far as the conventional fixed satellite service is concerned.
For conventional FSS communications links at C- and Ku-band, then, the available antennas are generally satisfactory, and advances are unlikely that would have major impact on cost or performance of satellite systems. The situation is different for mobile applications and ISLs.
With today's vehicular antenna technology, all proposed mobile satellite systems will be relatively starved for signal power, even with the largest foreseeable space antennas. Thus the availability of cheap, high gain, automatically tracking vehicular antennas would have a big impact on throughput and capacity. Mobile systems require spacecraft antennas that are capable of generating multiple spot beams and reconfiguring them quickly in response to changing traffic conditions. The necessary space and ground antennas are difficult to make at L-band, where both the GEO satellite mobile systems (American Mobile Satellite Corporation (AMSC), Inmarsat, etc.) and the "big LEOSAT" systems (IRIDIUM, Constellation Communications, etc.) will operate and at VHF, where the "little LEOSAT" systems (ORBCOMM, STARSYS, etc.) will operate, because the wavelengths are significantly longer than at C-band and the antennas must be correspondingly larger. To meet these requirements there is considerable interest in making phased arrays for both space and ground mobile satellite antennas and in developing large space antennas that will allow beam forming and beam switching techniques developed for C-band to be used. Restrictions on the physical size that practical launch vehicles can accommodate require that large reflector antennas be space-deployable.
Microwave ISLs for communications and data relay satellite systems share many of the same requirements with mobile systems. In addition, they impose significant scan requirements on spacecraft antennas. If met by physically moving large reflectors, momentum problems can created for the spacecraft. Several groups in Europe and the U.S. are looking at other approaches (direct radiating phased arrays, phased array feeds for reflectors, reflectors with mechanically scanned subreflectors).
In summary, the major unmet antenna needs for commercial satellite communications are for better vehicular antennas and for large space antennas with beams that can be scanned and reconfigured on demand.
Space antenna development observed in Europe was mission specific; typically, satellites used physically separate antennas for mobile service, ISLs, and broadcasting. Japanese space antenna systems are significantly more integrated; for example, a single 2 m reflector on the Communications and Broadcasting Engineering Test Satellite (COMETS) spacecraft supports Ka-band and millimeter wave (mm wave) mobile service, Ka-band ISLs and feeder links, and TT&C (tracking, telemetry, and command). This integration makes it difficult to provide a parallel discussion of Japanese and European antenna technology without repetition. We have attempted to solve it by describing two upcoming Japanese satellites with integrated antenna systems before discussing antennas for particular services.
ETS-VI. The Engineering Test Satellite (ETS-VI) (Kadowaki et al. 1992a, 1992b), scheduled for a 1994 launch, carries a payload which includes an S-band ISL experiment (SIC), an O-band communications experiment (OCE), a fixed and mobile communications experiment (FMC), a bidirectional space laser communications experiment (LCE), and a K-band single access ISL experiment (KSA). ETS-VI (Figure 2.8) carries a 7.5 cm telescope and five antennas: an S-band phased array, a 0.4 m reflector for 43/38 GHz, an 0.8 m reflector for 26/23 GHz, a 3.5 m reflector for 20 GHz and for 2.5/2.6 GHz, and a 2.5 meter reflector for 30 GHz and 6/4 GHz.
Figure 2.8. The ETS-VI Satellite
COMETS. The COMETS spacecraft (Isobe et al. 1992; Ohkawa et al. 1992), shown in Figure 2.9, will be launched in February 1997 to support experiments in Ka-band broadcasting, Ka- and S-band ISLs, and Ka-band and millimeter wave personal and mobile communications. The satellite will carry three communications antennas: a 3.6 m reflector for Ka- and S-band ISLs, a 2.3 m reflector for advanced Ka-band broadcasting, and a 2 m reflector for Ka-band ISL feeder links and mm wave and Ka-band mobile communications experiments. The Ka-band uplinks will operate at 30.75 to 30.85 GHz and the downlinks will be at 20.98 to 21.07 GHz. The mm wave uplinks will be at 46.87 to 46.90 GHz and the downlinks will be at 43.75 to 43.78 GHz. The broadcasting downlinks will be at 20.7 Ghz with feeder links at 27.3 and 27.8 GHz. (The COMETS frequencies, selected prior to the 1992 World Administrative Radio Conference (WARC 1992), do not conform to the WARC recommendations.)
Figure 2.9. Conceptual Sketch of COMETS
European Flight Programs. European Space Research and Technology Center's (ESTEC's) work on mobile satellite systems began with studies conducted by Alcatel Espace and Marconi under the ARAMIS program. Marconi featured a 2.2 m aperture direct-radiating phased array with 21 subarrays generating 6(superscript þ) beams for regional coverage and a global beam. A few beams could be scanned; the rest were fixed. The beamforming was done at intermediate frequency (IF) -- an advanced feature at that time. The design was never built, but it influenced subsequent work on the experimental ITALSAT F2, on the LMM payload for the ARTEMIS satellite, and on the commercial Inmarsat III family of spacecraft.
Much current R&D on space antennas for European mobile satellite communications is part of the European Mobile Systems (EMS) package for ITALSAT F2 (1994 launch) and the projected ARTEMIS (Advanced Relay and Technology Mission) L-band Mobile Payload (LLM) (late 1995 or early 1996 launch). EMS will replace the 40/50 GHz propagation package carried by ITALSAT F1. At the time this report was written (1992), the configuration of ARTEMIS was unresolved, but ESA intends that a subset of the communications characteristics of the ARTEMIS mobile package will be able to duplicate those of the EMS payload on ITALSAT F2. The ARTEMIS payload will be able to emulate the ITALSAT F2 EMS package and thus provide in-orbit backup for European mobile services. Alenia Spazio is responsible for both.
The ITALSAT F2 EMS system will provide single circular global coverage (Europe, part of Russia, and North Africa) transmitting and receiving beams at 1544 MHz and 1646 Mhz, respectively. The antenna gain is about 23 dBi and the effective isotropic radiated power (EIRP) is 46 dBW over the coverage area. (EIRP is the sum of the transmitter power supplied to the antenna and the antenna gain, in dBW and dBi, respectively.) The antenna consists of a deployable offset reflector 2 m in diameter, fed by a phased array with patch-excited cup elements.
The reflector is also used (with another feed system) to generate multiple Ka-band beams for Italian domestic service.
The ARTEMIS array-fed reflector antenna system is currently described as providing three or four beams (the original plans called for five) and an EIRP in the 48 to 49 dBW range. The feed network will be a four element phased array like that on ITALSAT F2, but the final reflector design is still uncertain. Its diameter is stated to be 2.85 m and it will be deployable. Apparently a final decision on the reflector configuration is awaiting word on the space available in the launch vehicle.
All the ARTEMIS mobile package hardware is under construction except for the antenna reflector. Breadboards exist for most of the electronic subsystems, the antenna feed, and the BFN.
Inmarsat III is a commercial satellite series scheduled for first launch in 1994 which is at least as advanced as the ITALSAT F2 and ARTEMIS mobile packages. The Inmarsat III antenna system will produce a global beam and five reconfigurable spot beams, and it is capable of continuously varying the power distribution between the beams to accommodate changing traffic conditions. The antenna itself is a 2.4 m reflector fed by an array of helical elements. The feed array can be rotated mechanically to reconfigure the beams.
Japanese Flight Programs. Japanese work in L-band space antennas for mobile satellite communications began with ETS-V, launched in August 1987. It carries a 1.5 m offset parabola with 26 dBi gain that provides two fixed beams, one covering Japan and the northern Pacific Ocean, and the other covering Australia and the western Pacific. With ETS-V, ETS-VI and COMETS, Japan will have gained significant experience in the design, construction, and operation of space antennas for mobile service from L-band through mm waves.
This section discusses space antenna technology applicable to satellite mobile systems that is not directly involved with scheduled flight programs.
Alcatel's concept for Inmarsat III employed a direct radiating active array for transmitting and a phased array feed with a reflector for receiving (both at L-band). The receiving antenna design is called FAFR for focal array fed reflector. The radiating elements are reflector-backed annular slots. Breadboard models were built, and the FAFR achieved an 80 MHz bandwidth at 1.64 GHz. It delivered left-hand circular polarized (LHCP) waves with a 3 dB axial ratio and 10 dB gain. (Note that this is the gain of the feed array and not the gain of the receiving antenna, the feed plus a reflector.)
The FAFR work stopped when Inmarsat did not accept Alcatel's proposal, but the technology developed reappears in an Alcatel proposed design for active and passive phased arrays for GLOBALSTAR. These must scan over a range of ñ55 degrees. Alcatel is planning a system with separate transmit and receive arrays at 2.5 GHz and 1.6 GHz respectively with 6, 8, 12, or 16 beams. The number of beams will be determined by the frequency re-use factor needed to obtain the desired traffic capacity.
Several European companies are actively developing large deployable space antennas with obvious application for L-band mobile systems. Thus MBB has produced a large, unfoldable L-band antenna for ESA as a technology demonstration. MBB has also collaborated with ANT in the design and construction of a 4.7 m unfolding petal antenna. A preliminary Alcatel design using a 5 m reflector fed by a 90-element phased array (Lenormand et al. 1988b) would develop eight spot beams, each with a minimum gain of 32 dBi.
Japan supports a major R&D effort in large space deployable antennas. Unfurlable antennas of the segmented foldable type are available now. Large petal, mesh, and robotically assembled antennas are being developed for big mobile communications satellites (Figure 2.10) or enormous geostationary platforms that would provide a variety of communications services (Figure 2.11 ). The Japanese expect these antenna technologies to be routine in five to ten years.
Folding Petal Antennas. In addition to the deployable ETS-VI and COMETS antennas described above, Toshiba has developed a 5 m Cassegrain antenna (Figure 2.12) that works at both Ka-band (23 and 32 GHz) and S-band (2.1 to 2.3 GHz). Its 24 petals fold into a 1.6 m diameter structure for launch; they are deployed by a single drive motor. The reflector has achieved +/- 0.35 mm rms surface accuracy, including thermal distortions. Gain is 58 dBi at Ka-band and 36 dBi at S-band.
Figure 2.10. Conceptual Drawing of a Large Mobile Communications Satellite (Courtesy SCR)
Figure 2.11. Conceptual Drawing of a Large Geostationary Platform That Would Provide a Variety of Communications Services (Courtesy SCR)
Figure 2.12. 23/32 GHz Petal Deployable Antenna Built by Toshiba for NASDA (Courtesy Toshiba)
Mesh Reflector Antennas. Developing large deployable mesh reflector antennas in the 30 m class is a project of the Space Communications Research Corporation (SCR). SCR is working on offset paraboloid reflectors to cover 800 MHz through 2.5 GHz with a surface accuracy of 0.05 wavelengths or less. Electrical issues are gain, cross-pol level, beam isolation, and passive intermodulation (PIM) products. Mechanical issues are deployment mechanisms, stowage, latching, surface accuracy, rigidity, and surface density. 3 m mechanical models using "hexa-link" (Figures 2.13, 2.14a) and 2.14b) and "tetrus" (tetra triangonal prism truss) (Figure 2.15 ) construction have been tested. A developmental 3 m tetrus-type C-band mesh reflector built by Toshiba with SCR sponsorship (Figure 2.16 ) has a mass of 16 kg; a scaled up 30 m version for UHF through S-band would have a mass of less than 700 kg. The expected surface accuracy over a -150 degrees to +100 degrees centigrade temperature range is ñ5 mm with active control by actuators. Adaptive control systems using integrated piezoelectric actuators to suppress vibrations and make precise adjustment to the truss geometry are being investigated (Shibuta et al.).
The MUSES-B satellite (the orbiting end of a very-long-baseline interferometer) will carry a deployable hexagonal 8 m diameter mesh reflector built by Mitsubishi Electric Corporation (MELCO) and operating at a number of frequencies between L-band and 40 GHz. The reflector is supported by a center post (which holds the feed) and six radial spars. The reflector doesn't rest directly on the spars but on a network of dielectric wires ("quartz fibers") that are connected to the mesh, to the spars, and to each other. No active control or in-orbit adjustment of the surface is possible or necessary. A 3 m prototype has been built. Its surface is highly reproducible from deployment to deployment (on the ground), and the surface accuracy is said to be good (no numbers were given). A 30 m model is being designed.
Robotically-Assembled Antennas. A Japanese program which, to our knowledge, has no counterpart elsewhere is the development of antennas robotically assembled in space (Suzuki et al. 1992). This is a "model mission" (possible experiment) for the Japanese Experimental Module (JEM) proposed for Space Station Freedom that has been studied at the Communication Research Laboratory (CRL) of the Ministry of Posts and Telegraphs (MPT) since 1986. Research has focused on the mechanical structure and on the latching mechanism. The design is assembled from a central section and petals using a charge-coupled device (CCD) camera for remote control. A 2 m prototype has been built and tested at 43 GHz with good results. Fujitsu is building the "dexterous hand" that will be attached to the ETS-VII robot arm for an on-orbit trial of all essentially mechanisms. The long-term goal is to move through the 2 to 5 m stage to the 10 m stage and ultimately to construct a 100 m antenna in space. Nippon Electric Corporation (NEC) and Toshiba are working on a robotically assembled Cassegrain antenna to be tested on the Space Station (Figure 2.17).
Figure 2.13. The Hexa-Link Truss Structure (Courtesy of SCR)
Figure 2.14a. A Mechanical Model of a 3 m Deployable Mesh Antenna Supported by a Hexa-link Truss Structure -- Stowed Position (Courtesy SCR)
Figure 2.14b. A Mechanical Model of a 3 m Deployable Mesh Antenna Supported by a Hexa-link Truss Structure -- Deployed Position (Courtesy SCR)
Figure 2.15. The Tetrus Structure (Courtesy SCR)
Figure 2.16. A 3 m Space Deployable Antenna Built by Toshiba/SCR (Courtesy SCR & Toshiba)
Figure 2.17. A Robotically Assembled Antenna for the Space Station Freedom Japanese Experiment Module (Courtesy Toshiba)
KOMETA has designed a 30 m unfurlable antenna (with accompanying stabilization system) which may be launched as early as 1996 on a large communications satellite operating at L-, C-, and Ku-bands. The associated beam forming network will generate 84 beams. Most of these will be fixed, but a few may be movable or scanning. This antenna is part of a system that could support five million 64 kbits/sec links to hand-held radios on the ground.