• The African Demand For Satellite Equipment
• Australian Satellite Market
• Asian Transponder Demand
• Multimedia Satellites - The Ka-band Report
• A Concept of Advanced Technology Satellite Systems for Asia Pacific: Rural
  Communications Applications

About Ka-band Multimedia Satellites

The Ka-band Report aims to provide a comprehensive survey of the current environment for Ka-band satellites. These represent the next major opportunity for the satellite communications industry.

The term Ka-band satellite is now generally recognised as a shorthand term for a new generation of communications satellites that will use on-board processing and switching to provide full two way services to and from small earth stations comparable in size to today's satellite television dish. To do this efficiently they will use multiple pencil like spot beams. A number of proposals also include use of inter-satellite links. A variety of orbits are being considered. Apart from the conventional geostationary orbit, both low earth and middle earth orbit systems have been planned.

Such Ka-band satellite systems have also been described in other terms such as "multimedia satellites", "ATM satellites", "broadband switched" and "broadband interactive satellites". The first two terms are generally inaccurate in that Ka-band satellites can be used for other applications than multimedia or providing an ATM platform. Indeed, it remains questionable whether many of the proposals are suitable for ATM. The Ka-band Report uses the most commonly used shorthand, Ka-band satellites" to describe such systems. The term "ATM Satellite" is also incorrect because on-board switching can involve either ATM switching or circuit switching or both.

Moreover, there are in existence a number of Ka-band satellites that do not use on-board processing or pencil beams or inter-satellite links and whose use has either been confined to experimental work or conventional telecommunications or broadcasting traffic. The Ka-band Report covers these satellites only insofar as they affect the new generation of satellites.

However, use of on-board processing and switching are not confined to Ka-band satellites. Already the satellite communications industry is beginning to think about using even higher frequencies to provide services similar to or beyond those planned for the new generation of Ka-band satellites. The Ka-band Report draws attention to Motorola's M-Star and Celestri projects.

Indeed, on-board processing and switching (effectively the provision of the equivalent of a sophisticated telephone switchboard on a satellite) are already employed in satellites providing mobile communications to hand held receivers. One of a new generation of mobile satellite systems (Iridium) employs Ka-band communications for links between fixed ground stations that interconnect the public switched telephone networks to the satellites. However, the links between the satellites and the hand held receivers are in L-band. A number of GSO MSS systems under development also use OBP and switching.

Understanding the relationship between inter-satellite links, pencil spot beams and on-board processing and switching is essential in understanding the commercial role these satellites are aiming to fulfil.

In effect, the switching capability is capable of making them operate like a public telephone network but with the facility to offer digital services with a wide variety of bit rates. Users will be offered ‘bit rate on demand’ and its variation, ‘variable bit rate on demand’ - that is to say they will only pay for the time that they use a link. This contrasts with conventional satellites where users usually have to pay for permanent leases. That only makes it economic to use satellites where there is a massive amount of information to be moved, such as TV channels and trunk telephony links.

In contrast, the Ka-band concept offers the equivalent of a local drop telephone circuit where the user pays for temporary lease of time. The Ka-band environment also allows an alternative, charging per bit of information moved.

However, such services require a lot of bandwidth because each link will usually be operating on a point to point basis. The prime method of using the available spectrum efficiently is to use multiple "pencil" spot beams, each covering only a small area of the earth. This allows frequency re-use in much the same way as a cellular phone network re-uses spectrum.

Use of multiple spot beams in itself demands the use of on-board processing and switching to direct each transmission path between the different spot beams used for uplinking and downlinking. Similarly, inter-satellite links with similar Ka-band satellites will demand switching capabilities.

In effect, the on-board processing and switching capability is a network management facility. The Ka-band satellite operators will each become like telephone companies but with their telephone exchanges in the sky.

In the long term it looks likely that there will be a "merger" of services offered by mobile satellite systems with those offered by "mainstream" Ka-band satellites.

The sudden interest in Ka-band satellite communications during the middle of this decade is not an accident. It comes as a result of supply economics meeting a perceived demand for new services.

On the supply side, nearly two decades of R&D into using the Ka-band for satellite communications has produced results. Hitherto, the satellite communications industry has been wary of using Ka-band because it is subject to substantial interference from rain. That problem has now largely been circumvented. Much more significantly, NASA launched an advanced Ka-band satellite in 1993 that uses on-board processing and switching, pencil spot beams and the Ka-band - the three key technologies for the new commercial Ka-band satellites.

NASA's ACTS satellite performed extremely well; the project was very well managed and involved testing a wide variety of new ground stations and services (the ACTS experiments are continuing).

By the mid-1990s it was clear that the existing frequency bands being used by satellite operators had insufficient spectrum left to provide new bandwidth hungry two way broadband services. The Ka-band, in contrast, was little used. The Ka-band available to satellite operators involved a massive 2.5-3.5 GHz of spectrum - that is to say some 4-7 times that available to some C-band satellite operators.

On the demand side, the growth of Internet as a mass-market consumer (and business) service and perceived likely growth in demand for multimedia (services combining two way video, voice, data and graphics) indicated a need to provide substantive infrastructure for two way services and associated switching technology. Ka-band satellites offered one major solution to meeting the demand.

Indeed, there is growing belief that the main market for Ka-band satellites will be the provision of high speed Internet and Intranet access.

Thus by 1997 some 59 Ka-band projects had emerged world wide alongside an intensified R&D effort to refine Ka-band satellite and associated technologies.

The Ka-band Report provides coverage of both these Ka-band satellite plans and the associated R&D initiatives.

Ka-band satellites represent a technology breakthrough. As is the case with so many such breakthroughs in the past, this will change the way things are done and in ways that are difficult to predict and even more difficult to forecast. In the earliest days of the telephone it was believed that its prime use would be to provide a form of cable TV home entertainment by allowing users to listen in to concerts.

The satellite industry is a lot more sophisticated in researching potential markets than the early proponents of the telephone. Nevertheless, Ka-band satellites appear to be pitched at serving a variety of niche markets, each with uncertain demand. With the possible exception of high speed Internet access, no "killer application" for Ka-band infrastructure is foreseen by the industry.

Some of the major players in the Ka-band environment are now undertaking the costly and detailed market research on which to centre their business plans. It is not an easy exercise to do; plans involving global satellite networks involve researching national markets throughout the world. Ka-band offers the prospect of new services in new markets, generally recognised as one of the highest risk market entrance strategies.

Similarly, the entrance of Ka-band into the communications marketplace is paralleled by the introduction of complementary technologies for which the demand is uncertain. The most notable of these are ATM and digital satellite television. Similarly, other new delivery infrastructures are being developed that would compete with (and sometimes complement) Ka-band satellite systems. There too, future market demand is difficult to forecast. Such competing delivery infrastructure includes interactive high-speed digital cable networks, ADSL and VDSL, LMDS and PCNs.

Moreover, it looks highly unlikely that the first Ka-band satellites will be in operation much before the year 2000. Any realistic forecast for the demand for Ka-band services will necessarily be long term and therefore subject to wide margins of error.

Despite these difficulties, The Ka-band Report has analysed each of the major service markets that Ka-band satellites may address and come to some robust conclusions.

One conventional piece of wisdom is that Ka-band will be initially rolled out either to service integrated Ku-band/Ka-band receivers to provide consumers with a combination of one-way broadcast and two-way interactive offerings or it will be targeted at big ticket corporate users.

In the consumer environment, The Ka-band Report concludes that stand-alone Ka-band terminals, without DBS TV capabilities, are much more likely to serve a mass market. That in turn suggests that aiming at corporate users may not be the most desirable initial market entrance strategy.

Other potential markets that The Ka-band Report considers include tele-medicine, tele-education, voice, local television, VSATs, "home-use VSATs" and satellite newsgathering.

However, all of these developments are dependent on the global marketplace for satellite communications being opened up. The reader might assume that because satellite television is now ubiquitous in the developed world and rapidly becoming available in the developing world that this is already the case.

Unfortunately, this is not the case where two-way satellite communications is involved. Any cursory knowledge of the satellite PCS environment will indicate the difficulties of gaining even limited market access outside of the country of origin of the satellite operator.

The PCS operators have used a variety of levers to gain access. These include lobbying at the ITU for spectrum, lobbying the US Government and the FCC to negotiate on their behalf in such fora as the ITU and WTO, offering franchises to local telecoms companies to retail their services and sub-contacts to local satellite and hand-held set manufacturers.

Such an approach is now being taken in the Ka-band environment. Teledesic appears to have approached just about all the world's telephone companies and satellite manufacturing and R&D establishments offering partnership deals. It has also lobbied national governments and the European Commission, "hijacked" the WRC'95 Ka-band negotiations and tied up a lot of now relatively scarce Ka-band spectrum. Teledesic is not the only Ka-band player to have taken this route.

The Ka-band Report details and analyses the key developments and issues in this geopolitical environment, including the lead role being taken by the US Federal Communications Commission, its DISCO II arrangements and the freedom (or otherwise) of non-US Ka-band operators to enter the lucrative US market, the concerns of the European Commission and European industry and World Trade Organisation negotiations.

Finally, in presenting an over-all picture of the market for Ka-band satellites, The Ka-band Report takes a special look at one market that is significantly different - Japan.

The research for The Ka-band Report was undertaken between February 1996 and January 1997. A revision, in the form of the second edition, was undertaken in June 1997.

It was clear that over this period the number of potential players in the marketplace was relatively fluid. Some satellite operators looked at providing Ka-band and found that it was outside of their main core businesses; others put together ideas that remain on the back-burner and some plans looked to have been dropped because of mergers and acquisitions. However, in general, the identified number of players and plans increased considerably and there was also a significant refining and development of thinking.

Management Summary

The next big development in satellite communications will be broadband interactive services involving a combination of Ka-band (or higher) spectrum, "switchboards in the sky" and inter-satellite links. The market for services using such technological features is unproven but currently looks to be a series of niches with no clear killer application. In the long term there is likely to be some form of integration of such services with narrowband satellite delivered PCS services. These are the main conclusions of a new survey of Ka-band satellite communications called "The Ka-band Report".

The Big Picture Business Environment: The development of such services is intimately bound up with the opening up of the international marketplace for information technology and the convergence of computing, broadcasting and telecommunications technology. The Ka-band satellite concept offers the prospect of quick and scaleable (therefore economically realistic) roll out of advanced infrastructure and services within this environment and is therefore an issue now high on the political and world trade agenda.

The Ka-band environment also involves a fundamental shift in the structure of the satellite communications market from one dominated by a combination of governmental sponsored international satellite organisations and regional and domestic satellite systems to global commercial satellite operations financed through the marshalling of vast investments from the international financial markets.

The USA Dominates: The current arrangements are dying because Intelsat and other ISOs are no longer perceived in the United States as suitable vehicles for maximising US interests in the satellite communications marketplace. Indeed, there is no doubt that the United States has taken the lead in both Ka-band technology and the drive to open up the international marketplace for satellite communications and therefore the interests of its own domestic manufacturing and communications industries.

By September 1995, US business was forced to declare its hand in planning to develop Ka-band satellite infrastructure. The FCC has set a deadline for filing for authorisation for such systems. What was revealed was a series of projects that involved global satellite systems with a high degree of vertical integration between satellite manufacturers and operators as well as a similar number of filings to protect the position of operators or of an opportunistic nature.

Since then the rest of the world has followed on with a series of plans for Ka-band systems, some of which are little more than paper plans to secure potentially scarce spectrum. Nevertheless most of the world’s satellite operators now have plans for Ka-band systems and at least three European satellite manufacturers have plans that involve a degree of vertical integration with satellite operating.

Nevertheless, few of the plans originating from outside of the United States involve true global networks; the mind-set still involves domestic or regional satellite operations. Only two non-US plans look to be truly global in capability (those from Alcatel Espace and SES).

This is a sign of weakness in competing with US interests. The United States has the upper hand in other ways. It is dominating the negotiation procedures on spectrum and opening up markets through the carrot and stick of allowing access to US domestic communications markets of all sorts. It is ahead of the rest of the world in developing the key technologies that will be used in the Ka-band environment - on-board processing and switching, inter-satellite links, phased array antennas and a vast array of other applicable technology.

The US Government and its industry are also twisting the arms of its potential rivals in the rest of the world. A key approach to developing its global satellite networks has been to "offer" partnerships to local telecommunications carriers and manufacturing firms. This both spreads commercial risk and assures that such partners have a vested interested in opening up the markets for the global satellite systems. It looks possible that other tools may be used to strengthen the market position of such global operations, such as allowing access to inter-satellite links.

Lessons for the Satcoms World: Clearly potential operators of Ka-band satellites are going to have to enter into partnerships with industry and communications carriers to develop their services and markets. Whilst it is theoretically possible that the WTO agreements will give the operators a free hand to enter any market they want, this is a politically and commercially naïve option.

The Ka-band technology offers the possibility of a direct attack on the biggest communications market of all, telecommunications (10 times the size of the broadcasting marketplace).

Satellite operators have hitherto concentrated on marketing transponders and bandwidth (usually through full time leases) to a limited number of broadcasting or specialised telecommunications customers, usually in limited geographic territories. They will need to adapt to a very different environment of selling variable bandwidth on demand and exploiting the value added of on-board processing and switching.

Their customer base will radically alter to a much broader base of commercial networks, intra-company communications, consumer markets (the US DBS operators are in a strong position here but not their equivalents elsewhere), interconnection arrangements with PSTNs, PCS networks, MSS infrastructure, inter-satellite links with other satellite operators and so on.

It is highly questionable whether the typical satellite operator of today, employing perhaps around 120-300 staff, is anywhere near experienced enough to undertake this alone and in a hostile and competitive international marketplace.

The Ka-band satellite operators will need to address the options of whether they want to remain wholesalers of capacity or full service providers (or a combination of both).

In either case they will need to partner with local telecommunications firms to gain market position, probably interconnection with PSTNs, facilitate obtaining licences and freedom to transmit data across borders, contact mechanisms to lobby politicians and so forth. Such local arrangements allow for mass marketing of Ka-band services that the satellite operators are ill-equipped to deal with.

Such partnerships also "buy out" potential competition. The concept can be extended to satellite manufacturing in a world where such manufacturers are also integrated satellite operators.

Who Wants the Technology? Serious players in the Ka-band market are now undertaking detailed market research into the likely demand for services. However, it is clear from conversations with industry players that there is no single killer application apart from high speed Internet access. Instead there is likely to be a series of niche markets ranging from ATM based multimedia for businesses and personal use, videoconferencing and video telephony, data broadcasting, voice for remote rural areas, tele-medicine, tele-education, SCADA, local television, a variety of applications in the mobile environment, satellite data relay services, news on demand, inter-connection of satellite and terrestrial PCS networks and satellite news gathering.

What this means in practice this means is that the Ka-band environment can offer sophisticated next generation VSAT type services including what the Japanese have called Home-use VSATs.

The one area that has been over-hyped in the Ka-band environment is the possibility of combining service with digital Ku-band DBS TV. In this scenario, a consumer could have a single dish that would provide both conventional TV and a plethora of two-way services, including, incidentally, voice based on distance insensitive tariffs.

The Ka-band Report concludes that this is unlikely to be a significant market because digital DBS is a niche market that only marginally overlaps with the likely market at consumer level for two way services. Bandwidth limitations for Ka-band preclude its use for mass-market video on demand. Provision of local TV through Ka-band to the DBS environment is likely to be viable only in the specific circumstances of the US marketplace and will only be viable if the DBS operators go into head on war with the cable operators.

Internet Access: Current thinking amongst many Ka-band proponents suggests that high speed Internet and Intranet services will be the major market driver. Ka-band satellites might be able to capture around 10-15% of the global market for these transmission paths. Commercial and public sector R&D organisations as well as carriers are now increasingly focusing technical and market research on this sector.

Constraints on Ka-band: There are some severe constraints in matching the supply side with demand.

In most cases transmission costs via satellite are likely to be significantly higher than those available where there is a reasonably sophisticated terrestrial infrastructure. That means in most of the developed world, where such infrastructure is not available, low levels of purchasing power will dampen demand. Perhaps the one clear exception to this is in the provision of high speed Internet access where the main alternative currently remains as ISDN and where the satellite option looks to be cheaper than using high-speed cable modems.

The use of geostationary satellites creates significant technical and commercial disadvantages where voice and two-way video are involved. This may rule out delivery of some services based on the ATM platform. However, there is now a growing interest in using low earth orbit and middle earth orbit to overcome the latency problem.

Use of LEO and MEO satellite configurations involve substantial initial capital investments by the satellite operators in contrast to the GSO option. Typically LEO solutions require a near complete global network of satellites before full commercial service can start. GSO requires only a single Ka-band payload on a single satellite before service can start and investment can be scaled up as and when demand increases. Typically the GSO environment may require initial investment (and therefore market exposure) of under $200 million whereas the LEO or MEO environment involves initial exposure well in excess of $1 billion.

The on-board processing and switching payloads are heavy, placing a severe constraint on what else the satellite can carry. In the engineering trade-offs needed to accommodate the OBP payload this may include reduction of station keeping fuel or increasing the mass of the satellite with associated increases in platform and launch costs. The R&D establishment continues to work on developing lower mass OBP equipment but development times are measured in years rather than months so the benefits from such work will stretch well into the next decade.

Costs of two way ground stations are far from clear, Whilst Ka-band proponents have suggested unit prices of $1,000 or less, there remains considerable scepticism whether such figures can be reached within the next few years. One potential Ka-band operator has suggested that the costs of ground stations will need to be subsidised. The use of LEO and MEO satellite configurations will call for the use of phased array antennas, which are significantly more expensive options than the conventional parabolic dish.

There is a growing belief that within the Ka-band there may be insufficient spectrum to meet the demand for services. Whilst it was initially expected that some 2.5-3.5 GHz would be available for GSO satellites, this is being whittled down by demands from spectrum for LMDS, MSS feeder links and non-geostationary Ka-band FSS satellites. The alternative is to go to even higher frequencies but their use in the commercial satellite environment is unproven and considerable expenditure on R&D may be required. This would delay start of services.

The Ka-band satellites are complex and require large platforms. Thus construction time is significantly higher than for conventional C-band and Ku-band satellites. It appears that the minimum time is about three years compared to 18 months now generally on offer by the satellite world. There is also some industry concern about the reliability of on-board processing systems that needs to be addressed by the satellite manufacturers.

There are no agreed standards shared between the potential operators of Ka-band satellites. In general the approach has been to let the market decide the de facto standards. Whilst there are strong arguments in favour of this approach (because it allows innovation and, potentially, a return on investment in developing equipment and intellectual property), the potential downside is loss of economies of scale and scope in manufacturing the ground segment.

Capacity Constraints Rule Out VOD: There are some significant limitations on the capabilities of Ka-band systems. In the VOD environment using current MPEG-2 based compression techniques a typical HS-601 satellite could provide around a maximum of 2,300 video circuits. This might be increased by a factor of 10 through a combination of more advanced compression techniques such as Wavelet technology and larger satellites. However, it is not a realistic option for mass market VOD.

Likely Rollout Timetable for Ka-band Satellites: Initial rollout of service is likely to be in a non-switched environment, using a limited number of transponders for bent pipe or data broadcasting type services. SES is planning to roll out services in 1998 at a time when Eutelsat could also offer ISIS type services using a combination of Ku-band switching and the Italsat Ka-band platform.

Most of the potential key players in the marketplace are secretive about their Ka-band plans but both Asiasat and Koreasat look to be front runners, with new Ka-band capacity available from 1999. None of the US proposals has yet involved an announcement to cut metal but one key player has suggested that it may initially roll out service for the Asian marketplace.

Most of the big players are looking to find partners to invest and develop service; so far there is little public confirmation that potential partners are committing serious investment. The typical reaction from PTOs is "we are considering all options". They are able to choose between partnering with the new fully private sector ventures such as SES, Cyberstar or Galaxy (to name a few) or to operations in which they already have an investment (domestic or regional satellite operators or the ISOs). The outcome is likely to involve a major strategic decision both for the PTO and the satellite operating industry.

It is therefore unrealistic to assume that major Ka-band platforms will be in operation until well into the year 2000.

Given the need to marshal major investment resources and commercial partners and the current still restrictive trade environment even this is optimistic.

Timetables have already been slipping - at one stage it was expected that construction of the first satellites carrying the Spaceway payload would begin in late 1995. Even though Hughes now has an international licence from the FCC and there is hearsay evidence to suggest it has lined up partners, there has been no announcement of go-ahead from the company.

Development of US domestic Ka-band satellites has also been delayed by the lack of award of licences from the FCC - it has faced difficulties in getting together a workable framework for the different claims on Ka-band spectrum and auctioning policies (the latter looks to have been dropped for Ka-band satellites).

Nevertheless, potential users of Ka-band satellites are now looking seriously at the opportunities open to them. Some of the satellite service providers have expressed the view that Ka-band is too far downstream for them to begin serious planning. Others, however, are now undertaking market research. The Ka-band environment offers them (and others) the possibility of offering value-added services over and above straight transmission. These include complete turnkey services using leased capacity (such as T1 circuits) to offer content to end users or complete telecommunications networks.

Developments over the last year in the US telecommunications marketplace also look good for Ka-band. Whilst the received wisdom until mid-summer was that the cable companies and the telcos were about to engage in all out warfare to enter each other's marketplace, there is now a stand-off as they realise costs do not match potential benefits. The consequence is that heavy investment in providing new services is not taking place on the scale envisaged. The cable companies are no longer rushing to provide two-way digital services and the telephone companies have cold feet about providing broadcast overlays.

Likely Structure of the Future Ka-band Marketplace: Whilst there are some 59 planned Ka-band satellite systems, the vast majority of such plans will fall by the wayside. Ka-band is one element in an overall restructuring of the world’s satellite communications marketplace.

In the medium term (years' 2000-2010), the overall satellite communications marketplace (excluding MSS) will consolidate around major commercial international players (perhaps three to four systems) with a limited number of regional systems (notably in Asia). The domestic players will, for the most part, either die or be taken over or, where they are particularly strong, expand into regional operations.

The future of the ISOs in this environment is uncertain as there is now considerable pressure to ensure that their dominant economic power is curtailed.

In the long term (after the year 2010) a feasible scenario may involve a second-generation global network of switched satellites providing a combination of narrowband and broadband services through both geostationary and MEO or LEO satellites. Such networks could be meshed together with inter-satellite links.

It also looks likely that there will be no winner in the battle between LEO/MEO systems and GSO systems. Ka-band GSO infrastructure is already going into place, which means that a market will be locked into using it through fixed earth stations.

Nevertheless, the technical shortcomings of geostationary satellites leaves a major market for non-geostationary systems that may include provision of conventional digital broadcast services. Line of sight is an endemic and permanent problem for all forms of GSO spacecraft (MSS, DBS and FSS). MEO and LEO configurations can address the market gap unserved by GSO (indeed, highly inclined elliptical orbits remain a further option not so far seriously envisaged by the commercial Ka-band fraternity).

Europe Needs to Get its Act Together: From the supply side, the key three markets in the world involved in Ka-band are the USA, Europe and Japan. The European satellite communications marketplace is over dependent on conventional DTH/DBS TV. This is now a mature marketplace and the digital successor looks to be of very limited potential. Yet its satellite operators are committed to rapidly expanding capacity to serve the DST marketplace. SES has three satellites and a spare on order for this marketplace and Eutelsat has a further three Hot Bird satellites under construction. There are also two additional Scandinavian satellites going into orbit this year and which are primarily dedicated to TV traffic.

The only one of the European satellite operators that has positioned itself for the Ka-band era is SES which has a modest commitment in the form of two transponders. Eutelsat’s thinking on Ka-band is opaque to say the least and it is far from clear whether its PTO owners would fund investment in a regional Eutelsat Ka-band infrastructure.

Moreover, Europe’s plans for Ka-band satellite systems are focused on serving the low growth European marketplace rather than the global environment. The Asian environment is largely ignored but even the most cursory analysis shows that this offers both the largest and fasted growing market for satellite communications. Indeed, anecdotal evidence suggests that major US proponents of Ka-band see China as perhaps the largest Ka-band market of them all.

Europe largely stopped R&D on the Ka-band after the in-orbit failure of the Olympus satellite. There has subsequently been almost an emergency response to developments in the USA through funding of such projects as Secoms. Nevertheless, despite an active OBP programme in ESA, there are no firm plans from the organisation to launch an OBP payload and France has opted out from the ESA framework in favour of Stentor. At the moment it is envisaged that this will not carry a Ka-band payload and France is also having problems in funding the project.

Whilst Italy has considerable experience of the Ka-band through its Italsat project, Alenia Spazio is heavily dependent on EC funding for development of the Euroskyway project. Indeed it is believed that Alenia Spazio has asked for much more substantial EC funding.

What is under question is whether Europe has the technological base to compete effectively in the manufacturing of Ka-band satellites and whether its satellite operators will, in the long term, be key players in provision of Ka-band infrastructure.

Europe is also way behind the USA in the development of phased array antennas.

However, the European Commission’s coordinating "Action Plan" on satellite communications may produce some results. The European Space Agency is expected in the second half of 1997 to take further initiatives in giving European and Canadian firms contracts to develop multimedia satellite technologies.

The Peculiar Circumstances of the Japanese Marketplace: The Japanese market for Ka-band communications remains different from that of the rest of the world. It looks as if Japan is trying to jump one-step ahead of Ka-band to provide GBit/s levels of communications as part of a broader, and still relatively dirigist, approach to communications.

Japan has a long in place Ka-band infrastructure used to provide trunk telecoms capacity for the PSTN. It also has a substantial and evolving satellite R&D policy now refocusing away from satellite platform and payload development to a dual environment including tighter focus on provision of service.

The position of the key Japanese satellite operators in the broadband multimedia environment is unclear. JSC has not filed for Ka-band capacity but SCC has filed for a significant increase in its Ka-band capacity (it already has Ka-band capacity but with onboard processing and switching).

Is Ka-band a Will o’the Wisp? No. The commercial risks of the Ka-band environment are probably considerably lower than the LEO/MEO MSS environment but the latter has already seen massive marshalling of finance and is just about to start service. The Ka-band environment offers less risk because it addresses the needs of a wide portfolio of services and demand. It also facilitates development of a wide variety of new services and applications. The use of GSO satellites in the Ka-band environment also presents the possibility of considerably lower capital investment exposure than the MSS environment.

The current constraints on developing Ka-band are mostly technical or lack of access to markets because of trade constraints. Both are now being addressed.

This is not to say that some of the thinking behind offering Ka-band services is flawed. One conclusion of The Ka-band Report is that the Teledesic approach is extremely high risk - it addresses a narrow niche centred on PC to PC communications and Internet access and involves a capital exposure of at $9 billion. Moreover, the technical aspects of the configuration are questionable. It is highly unlikely that the satellite system can be integrated with other services such as MSS and DBS.

One also suspects that it is politically unrealistic given that hidden barriers to trade and protection of markets will be endemic over the next decade despite any WTO agreements.

There is some clear indication that Ka-band multimedia satellites offer major end user cost savings over conventional Ku-band systems. US consulting firm TelAstra has compared the costs of offering data communications through Ku-and DBS satellites and through Ka-band. It found that the former would involve transmission costs eight times higher than in Ka-band.

Risk Analysis: Basic analysis of the market positioning of satellite based switched services suggests that many new services will be high risk ventures to launch. Low risk launches usually involve existing products into existing markets; highest risk launches involve new products in new markets. The relationship between whether the product is new or existing and whether the market is new or existing can be plotted on a simple 2*2 matrix, with the highest risk opportunities lying in the bottom right hand quadrant. (Table not shown here)

The implication of inter-satellite links: The plans from US GSO operators aiming at the global marketplace indicate a potential shift away from current dominant orbital slots in the international satellite market place. Inter-satellite links obviate the need for satellites placed over the main Ocean Regions to provide international communications. Instead, operational satellites can (and are planned to) be centred over landmasses where there are customers and where better line of sight can be provided.

New dominant orbital slots will therefore emerge and content providers may be forced to use them (rather than existing slots) to gain access to households. Moreover, the combination of global coverage and inter-satellite links means that each content provider may only need one uplink site to provide global distribution through a plethora of spacecraft.

Nevertheless, it is difficult not to conclude that inter-satellite links could be used as a tool by some operators to leverage their market position into an anti-competitive environment. This is particularly the case if there are no agreed technical standards for inter-satellites and no regulation of inter-connect rates.

Key Developments Expected in Ka-band in 1997: Whilst the Ka-band opportunity featured highly in 1995 in the debate on the future of satellite communications, developments in 1996 were distinctly low key, if not inconclusive.

Nevertheless, 1997 will see some major milestones in the progress towards operational broadband multimedia satellites -

The WTO negotiations on opening up the market for satellite communications were completed on 15 February 1997. These may render the FCC’s DISCO II proposals on opening up the satcoms marketplace either semi-redundant- or the basis for future bilateral negotiations.

In early May 1997 the Federal Communications Commission finally granted licences to the geostationary orbit applicants following a log series of negotiations with the players involved.

On 27^th June 1997 the European Commission’s Action Plan on satellite communications was agreed by the Council of Ministers. Whilst the Commission has very limited economic and legal power in the satcoms arena, the Action Plan represents its attempt to coordinate European policy in the face of intense competition from US interests.

Further orders for Ka-band space segment are expected this year with Koreasat and Asiasat being front runners in Asia and SES in Europe.

The number of serious Ka-band plans emanating from the USA will reduce. Such consolidation is natural at this stage in market development.

There will be growing interest in the LEO and MEO Ka-band option and, possibly, the use of even higher frequencies.

Two-way Internet access services using Ku-band (and, possibly, C-band) will be rolled out through 1997 as equipment manufacturers and service providers instigate new products and services. These will be precursor services to Ka-band Internet delivery.

The ATM Forum will set standards for ATM standards for use with satellites.

In October and November WRC ‘97 will cover spectrum and Ka-band for feeder links and use of high frequencies as an alternative to Ka-band.

The major proponents of Ka-band will continue to seek partners amongst PTOs, other carriers and industry to finance and market their projects.

Table 2.1: Summary of Planned Multimedia Systems Worldwide

Source: DTT Consulting

Commentary on Table 2.2: This table is intended to summarise all the known plans for the next generation of Ka-band satellites world-wide. However, it relies heavily on both FCC and ITU filings; in both cases the filings are likely to involve requests for more orbital slots and spectrum that the satellite operator envisages using. There have also been substantial modifications to filings at the ITU, with some operators reducing the number of slots and spectrum requested for co-ordination. Moreover, there is no direct relationship between filings made at the FCC and corresponding filings for them made by the USA with the ITU.

Where it is possible, DTT Consulting has estimated the real number of satellites that are likely to be ordered to contrast with claimed plans. This is almost impossible to do for the USA without frequent access to print material available at the FCC.

In some cases no filings have yet been made with the ITU or have, for reasons unknown, been completely withdrawn (Dacomsat 4). However, in both cases we have assumed that these are real (albeit paper) plans.

It is also suspected that a number of other existing or proposed satellite operators are planning to make filings at the ITU for Ka-band. However, we have made no assumption that they will and have therefore excluded them from the table.

The list excludes R&D and demonstration satellites and existing satellite systems using Ka-band such as Italsat, DFS-Copernicus, NStar and Superbird. It also excludes Big LEOs using Ka-band for feeder links and the Japanese aeronautical/GPS and date relay satellite MTSat.

However, because the M-Star satellite system appears to be closely associated with Motorola’s Ka-band plans, this has been included in the table. Similarly SkyBridge as been included although it is a Ka-band project.

Table 2.1 indicates that use of some 279 orbital slots by 307 geostationary satellites is planned in addition to 497 spacecraft in non-geostationary orbits. In total, 804 satellites are planned to be launched. Excluded from the latter are in-orbit spares for non-geostationary systems. However, in-orbit spares for geostationary systems are included as they tend to be used for non-essential traffic.

The table also indicates that the number of networks planned is 59 and the number of existing or potential satellite operators involved is 55.