While the hundreds of satellites have been the symbol of progress in satellite communications, the millions of ground terminals in all frequency bands represent an equally profound achievement. The ground segments now comprise a large portion, if not the majority, of the total cost of a new system. Satellites and terminals have tended to be independently produced, suggesting that they are in fact independent (with notable exceptions). But in recognition of their significant percentage of system cost, the relationship of satellite system design and terminal design¾and ultimately terminal cost¾has become more openly recognized and debated. Design activities for personal and mobile consumer-oriented systems that demand low cost terminals are largely responsible for the increased visibility of this important principle; and, indeed, led to its observation during the WTEC panel's site visits.
Much cost oriented terminal design has been taking place since the 1992/1993 study, and is implicit in the maturation of mobile and personal systems. The same principle must be employed in the future for consumer and business user Ka-band multimedia systems. An aspect of the terminal design consideration is the need to exploit the cost versus quantity relationship (Brandon 1994).
The anticipated continued growth of VSAT networks and satellite television broadcasting, and the introduction of personal and mobile systems and direct-to-user-services, together provide a broad basis for anticipating a very large future market for satellite communications terminals. In general, cost will decline as a function of large quantities, lower frequency and smaller size. The original U.S. ground terminal at Andover, Maine employed a 177 ft. steerable folded horn. A folded horn resembles a cornucopia; and the Andover horn is, in fact, a beginning point or inspirational source for all satellite terminals. Figure 4.1 depicts the present and predicted future populations for five classes of terminals discussed in this section. This chart depicts the evolutionary trajectories in quantity, size and frequency space, flowing out of the Andover horn. Quantities reach into the millions for several classes. The figure summarizes both the actual history and a future projection.
Major trend-making terminal classes with future impact are discussed in this section. Several classes, including news gathering terminals, are not discussed.
Fig. 4.1. Evolution of satellite terminals in number of terminals, frequency and size (1965-2007).
Very small aperture terminals (VSATs) have extensive uses in business and government and are anticipated to have expansive roles in future Ka-band systems. VSATs are here considered to include a transmit capability. Small receive-only terminals are a separate category discussed elsewhere in this chapter.
In developed countries, the ability to bypass existing infrastructures with a private network has achieved cost savings. In less developed countries, the possibility of establishing a distance-insensitive, modest cost network, using a satellite transponder with VSATs and a hub station, enables many cost-effective applications. The broad generality of VSAT uses and applications will enable continued worldwide growth.
The majority of VSATs operate at Ku-band. The worldwide number is difficult to determine with any precision. We estimate that for 1997 there were about 300,000 units worldwide. Considering the needs in Asia, Africa and S. America, the number should easily double within ten years.
While Ku-band is allocated for fixed service, an airborne Ku-band VSAT exists and there has been much activity in developing airborne receive-only array antennas. Historically, C-band VSATs were allowed through introduction of spread spectrum signaling that prevented interference (due to broad beamwidth of C-band VSATs) to adjacent satellites. Similarly, OMNI-Tracs is a mobile (vehicular) service (at Ku-band in the United States and Europe, at C-band in Latin America). It may be that similar spread spectrum techniques will allow introduction of airborne VSATs in Ku-band.
The VSAT consists of an outdoor unit¾antenna with low noise amplifer (LNA) and power amplifier located at the feed to minimize line loss, and a MMIC low noise down converter¾and an indoor unit (down converter and digital electronics that vary depending on the application). With the advent of large quantity production, VSATs may be integrated from parts available from many sources. Much experience is needed for volume production of high reliability units. Figure 4.2 shows a group of VSAT outdoor units at Hughes Network Systems being operated to assure quality before shipment.
Fig. 4.2. VSAT terminal quality assurance test range (Hughes Network Systems).
A number of Ka-band systems are planned for providing wider bandwidth (>10 kbps to 100 kbps) and "wide bandwidth" (~1.5 Mbps to 155 Mbps) services to small VSATs (~45 cm to 60 cm diameter antennas). These systems are based on providing video conferencing, private "intranet" services, telemedicine, teleeducation, direct two-way Internet access and multimedia communications of the future. As described, the systems are a new type of fixed satellite service (FSS).
Many of the envisioned services and applications would be offered to the private or home user. There is broad consensus that the VSAT terminal for consumer application must be carefully engineered for a total cost of $1,000 or less. Systems providers are engaged in defining means for accomplishing this goal. While details are proprietary, it appears that the cost goal is within reach.
The business user terminal would be expected to access satellite capacity with higher burst rates and simultaneously serve multiple individuals, for example, at one business location or facility. Because of the business application, service to multiple users, and other factors, the business terminal would be somewhat larger and higher in cost. Planning estimates have suggested a cost goal of about $10,000 for the business terminal. The business Ka-band terminal could use many of the components developed and produced for the private or home use terminal, thereby realizing benefits of a larger production base. To reach the cost goals, application of the cost versus quantity principle must be recognized and creatively applied.
An example of a multimedia VSAT design which is innovative yet sensitive to cost is that under study by KDD (see site report for KDD, Appendix C). A bi-directional (transmit and receive) multimedia service is envisioned with a 46 cm aperture receiving a 40 Mbps time division multiplex (TDM) waveform transmitted by a 7 meter hub. The return link (from user to hub) would be 128 kbps binary phase shift keying (BPSK) chirped (for low cost) to spread the energy over a 500 kHz bandwidth.
Terminals for systems such as Teledesic that use lower altitude orbits introduce the problem of "handover" (from one satellite to another) during a session. Achieving low cost terminals would seem to require a single antenna aperture and rapid handover also suggests phased array antennas.
Because of the compelling nature of video, direct-to-user satellite delivered television is possibly the most important medium ever produced for mass communications; and mass communications is achieved only because of the low cost receive-only terminal (TVRO). Television receive-only stations now outnumber all other types of terminals, with highest populations in North America, Europe, and Asia.
Ku-band television receive-only stations are now proliferating in the United States, Europe and Japan. (Some of the factors for this pattern are outlined below.) The typical TVRO antenna is about 46 cm (18 inches) in diameter. A LNA followed by a block frequency downconverter produce an rf signal at a microwave intermediate frequency (IF) suitable for transmission through a coaxial cable. The low noise block converter has been mass produced in the form of a monolithic microwave integrated circuit (MMIC). This device has become the most widely produced microwave component in history. The IF signal is delivered to an indoor unit (usually termed a "set top box" in digital television applications). The set top box selects the appropriate carrier, processes the signal and converts it to analog form for presentation. Digital television has become economical due to compression algorithms that allow 10 or more television channels per carrier/transponder.
The technology for producing TVRO terminals is well established and costs have been driven down by mass production. Current world totals are shown in Figure 4.1.
Distribution of analog television signals (e.g., to cable head-ends and hotels; in education systems, sporting or other news events; and for major network program distribution) remains a major use of satellites. The population of terminals for this type of use is again difficult to estimate (but is likely to be on the order of 10,000 in the United States). Within the United States, availability of these downlink signals (at C-band) led to low cost "back yard" antennas for private use; the population of these is estimated to be 2.5 million, and is not likely to increase due to the availability of digital TVROs at a fraction of the cost of the larger analog units.
When digital broadcasting was introduced in Japan, the number of digital TVROs purchased reached a million within six months. The antennas are so small that they do not violate cultural concerns for clarity and order, and may be seen today throughout Japan. This is an important observation, because the multimedia home or personal terminal will be of similar size and configuration, and therefore will have no impediment to broad application. Some private TVRO antennas in use in Japan are shown in Figure 4.3.
The dream of a handheld terminal for satellite communications seemed distant as recently as 1990. But imaginative application of technology developed for use in other contexts produced system designs enabling handheld voice terminals.1 Handheld data terminals have also been created for little LEO data systems. Orbcomm terminal designs have been complete for several years. However, no handheld satellite terminals of any kind are as yet in volume production.
The handheld terminals for voice are being designed and produced for market trials beginning in 1998. Data terminals are not yet in volume production but are available as engineering models from multiple sources. The hand held data and voice terminals are exemplified by the Torrey Sciences data terminal for Orbcomm, and the Motorola Iridium handset, illustrated in Figure 4.4. A small transportable terminal for use with store and forward microsatellites (i.e., typically 50 kg in weight), both produced by the Surrey Satellite Research Center at University of Surrey, U.K., is also illustrated. (The store and forward link data rate with the Surrey microsatellites is typically 10 kbps, and about 750 kilobits can be received in a single satellite orbital pass.)
Fig. 4.3. Satellite television receive-only terminals in Japan.
The terminal concepts in the center row of Figure 4.4 are based on the ubiquity of personal computers and related technology. These concepts may use a computer as the input/output device (i.e., for composing messages and displaying received messages), becoming a satellite terminal by addition of a small applique box, similar to an "outdoor unit" for a VSAT; or the rf functions may be integrated to produce terminals that resemble laptop computers. Some of these terminals incorporate a voice capability and are highly portable but are not considered "handheld" for purposes of this discussion.
Globalstar handsets are termed user terminals or UTs. Reflecting the multi-mode philosophy, there are 3 types of UT: Globalstar only; Globalstar & GSM; and Globalstar, GSM & AMPS. Qualcomm is designing and building the handsets; Orbitel (owned by Ericsson) will build handsets in the U.K. in addition to suppliers in Italy and Korea. The CDMA parts are delivered by Qualcomm; the power amplifier chip is made in Japan. Currently the GSM/AMPS parts are joined with a CDMA phone, with no integration except for battery, microphone and headphone. Integration will follow later (functions on single chips, same data rates, etc.). UT software comes from terrestrial cellular phone applications. As is also true for Iridium, the Globalstar gateway software has taken major large pieces of code from terrestrial base station software.
The Globalstar UT is light and has a deployable, dual quadrifilar helix type antenna that must be held vertically. Doppler effect is compensated for between gateway and satellite. Predictions help the UT to acquire the signal quickly. Call setup is via a random access channel; after a connection has been established, all control information is transmitted via the communications channel, including the power control (update rate, order of seconds). The CDMA handset could benefit from miniature filter technology realizing lower out-of-band emissions and also from a more efficient high power amplifier (HPA). Filters for the out-of-band emission problem are difficult for CDMA, in particular for a higher power automobile unit. The ICO system is expected to produce a handheld voice terminal but details are not yet available.
Fig. 4.4. Handheld and highly portable communications satellite terminals.