16
Satellite Networks
16.1 Potential Convergence
Getting coverage to deep rural areas or out at sea or on deserted mountain tops or in the middle of the desert where no one lives can be technically challenging and economically problematic. With the possible exception of coverage in the middle of the ocean, cellular networks can provide coverage more or less anywhere, but the cost can become prohibitive relative to income.
Cellular networks are generally speaking optimised for capacity and increasingly now high per user peak data rates. The wider channel spacing and wider operational bandwidths needed generally reduce the coverage available in any particular band.
This chapter looks at low-earth orbit and geostationary satellite systems but we start with two-way radio systems and specialist user requirements. As with terrestrial cellular networks, satellite networks and two-way radio networks are increasingly doing a lot more than just supporting voice traffic and are becoming increasingly information centric and, in the jargon of the sector, ‘mission critical’.
Table 16.1 Specialist user needs
| Wide-area coverage | Better than cellular particularly in rural areas and/or within buildings in urban environments |
| ‘All informed’ user capability | The ability to hear other users in an ‘open’ channel |
| ‘Instant access’ onto available channels | Call setup time has to be less than 250 milliseconds |
| Multigroup announcements | |
| Wide-area broadcast messages | |
| Dynamically changeable priority levels | |
| Secure authentication and encryption | |
| Voice clarity | Includes task optimised noise cancellation, specialist hands-free and whisper phones for covert surveillance |
| Press to talk | With option to work full duplex if needed |
| Talk groups | Geographical or functional |
| Interworking/interoperability | Problematic as we shall see later |
| Storm plans/special event plans | Preset response to particular events including disaster recovery contingency planning. |
| Sleeper phones | The ability to stun phones remotely and re-enable them as listening or more recently listening and watching and sensing devices. |
| Ruggedised hand sets | Waterproof, dustproof, shockproof fireproof and explosionproof handsets |
At the application layer substantial standards commonality is beginning to emerge across all delivery systems, for instance the use of MPEG and JPEG encoding and compression schemes and MPEG-based content descriptors. At the transport layer IP protocols are becoming more pervasive and potentially allow much greater transparency across multiple radio bearers. However, the benefits of this transparency can only be realised if a closer commonality between physical radio bearers can be developed over time.
16.2 Traditional Specialist User Expectations
Specialist users include public safety agencies, fire, police, ambulance, defence users and nonemergency users in transport and distribution, energy and small to medium business users.
Traditional expectations in terms of user functionality include open channel working between multiple users based on push to talk functionality, back-to-back direct mode operation where users can talk to each other directly without a base station in between and ground-to-air/air-to-ground communication, particularly important for military users or event management where aerial surveillance is required.
Latency has always been important and the benchmark generally has been to deliver onto channel times of less than 250 milliseconds.
In terms of radio performance, the most important metric has traditionally been to deliver good rural coverage and good urban inbuilding penetration for two-way voice communication.
In Europe this has been achieved by working at VHF or UHF with both the handsets and base stations working at higher powers than cellular radio. The narrow channel spacing used, typically between 6.25 and 25 kHz and the relatively narrow operational bandwidths, typically 5 MHz or less, makes it possible to deliver good receive sensitivity.
In the USA there are substantial two-way radio system deployments in the 800-MHz band. In common with European systems these operate at higher power than cellular systems. APCO 25 portables, for example, are allowed to produce 5 watts and the base stations can produce up to 500 watts. APCO is unusual in that it is standard developed by a specific user community, the Association of Public Safety Communication Officials.
Table 16.1 summarises some typical traditional specialist user needs.
From the above, it can be seen that it is not always easy to move users from a two-way radio system onto a cellular system and provide them with sufficiently equivalent functional capabilities. The attempted migration of Sprint Nextel iDEN users to Sprint CDMA a few years ago is a legacy example of some of the challenges inherent in this process.
16.3 Impact of Cellular on Specialist User Expectations
Conversely, much of the functionality that we take for granted in cellular handsets, particularly multimedia functionality and the data rates needed to support the real-time exchange of multimedia content is either not available in a two-way radio or commands a substantial cost premium and/or weight and size and/or duty cycle penalty.
Not all specialist users need or want multimedia capabilities in their handsets, but some do and this present minority will likely increase over time. Multimedia functionality implies audio and video capture bandwidth using high resolution cameras and high resolution display bandwidth using high-resolution displays. These functions are only immediately useful if matched with efficient high data rate uplink and downlink radio bearers.
It is not impossible to deliver adequate data rates over two-way radio systems. There are wider-band implementations of two-way radio systems potentially capable of meeting multimedia specialist user expectations but substantial engineering effort and investment will be needed to make the handsets and networks work efficiently. The narrowband technologies are optimised for voice or low bit rate data so are inherently unsuitable for real-time multimedia. These narrowband systems are spectrally efficient but this is of marginal relevance to most specialist end users.
We can illustrate these practical data rate limitations by reviewing first the technologies used and secondly the spectrum into which those technologies are presently deployed.
16.4 DMR 446
This is a digital PMR standard developed through ETSI and called 446 as 446 MHz is where some license exempt spectrum has been made available in some countries, for example the UK.
It is neither intended nor suitable for multimedia transmission. DMR products are available from Motorola, Icom, Kenwood, Tait and Entel and could potentially be deployed into other UHF and particularly VHF PMR applications. As we shall see later there is a problem of scale economy with these products and their target markets.
16.5 TETRA and TETRA TEDS
TETRA is originally a European standard for trunked radio based on 25 kHz channel spacing with a 4 slot frame structure. A more recent extension of the standard, TETRA Enhanced Data Service (TEDS) scales the 25 kHz channel spacing to 150 kHz and uses 16-level QAM to support up to 500 kbit/s as a ‘close to cell’ data rate.
TETRA handsets are available from Motorola, Kenwood, Sepura and Selex. Nokia has not developed handsets for this market since 2005. Even assuming general availability of TEDS complaint handsets, present TETRA cell densities would mean these higher data rates could only be supported in limited geographic areas.
This is not presently a major problem. Many specialist user communities are rightly hesitant about adopting new technologies or new variants of existing technologies. The sale of the TETRA Airwave network in the UK in 2007 for close to £2 billion pounds proves that good revenues and margins are available based on the provision of voice combined with relatively low data rate services. However, it is reasonable to expect that this will change over time particularly if the performance gap between TETRA radios and cellular handsets continues to widen, which it will for reasons that we explain later.
16.6 TETRAPOL
The ‘other’ European trunked radio standard introduced by Matra, now EADS, based on narrowband 12.5 kHz channels using GMSK. Tetrapol delivers very good range when used for voice services but is bandwidth limited for multimedia.
16.7 WiDEN
WiDEN is an evolved version of iDEN, the Integrated Digital Enhanced Network standard developed by Motorola in the early 1990s. The frame/slot structure is either 3 slot or 6 slot similar to cellular US TDMA. When combined with 64-level QAM modulation this can deliver per user data rates in the order of 100 kbit/s, but no WiDEN products are presently available.
16.8 APCO 25
APCO25/Project 25 handsets have 12.5-kHz or 6.25-kHz channel spacing and a one- or two-slot air interface using QPSK. The handsets have good range performance but are bandwidth limited for multimedia. Manufacturers include Motorola, Tait, Maxon and Bosch.
As can be seen, none of the available two-way radio technologies summarised in Table 16.2 are ideal for multimedia. The standards that are capable of supporting the higher data rates needed do not yet have products available.
Table 16.2 Comparison of present two-way radio technologies
16.9 Why the Performance Gap Between Cellular and Two-Way Radio will Continue to Increase Over Time
This is not specifically a technology constraint but more a consequence of limited engineering resources. Substantial engineering resources are needed to develop and ratify standards.
Substantial engineering resources are then needed to interpret these standards and translate them into performance-competitive cost-competitive radio products, handsets and base stations.
This is one reason why the performance gap between cellular phones and two-way radios continues to widen. Silicon vendors and handset manufacturers producing cellular and mobile broadband quad and quintuple band chip sets are servicing a unified market that is now running at over one billion units per year.
Companies like Nokia and the silicon and component vendors supporting Nokia and other Tier 1 cellular handset manufacturers have to treat this dominant market as an absolute priority when deciding on the allocation of engineering development resources. This ‘market pull’ gravitational effect also helps to focus efforts on meeting other user requirements such as interoperability and roaming.
The opportunity costs of servicing markets that are several orders of magnitude smaller compounded by the ‘divide down’ effect of needing to support multiple standards for these minority markets makes it extremely difficult to justify investment in two-way radio handset or infrastructure radio hardware and software development.
The result is that products come to market more slowly, are more expensive, and are generally more limited in terms of their radio functionality, at least as far as overall data rates are concerned.
16.10 What This Means for Two-Way Radio Network Operators
So this tells us two things. Specialist user expectations are changing over time. These expectations are partly driven by personal exposure to rapidly evolving cellular handset form factor and functionality.
These expectations include the assumption that multimedia capabilities can and should be made available in small form factor portable devices. This expectation extends to include the assumption that multimedia capabilities should be available whenever and wherever voice service is available.
This is presently a major challenge for cellular service providers but is an even greater challenge for two-way radio network operators with networks designed and dimensioned for voice and standards that have been historically driven by voice performance metrics.
16.11 Lack of Frequency Harmonisation as a Compounding Factor
Two-way radio hardware development might be more attractive if there was at least some degree of global commonality in terms of band plan allocation.
Table 16.3 shows a rather over simplified representation of present European VHF two-way radio and radio/TV allocations
Table 16.3 European VHF two-way radio and radio/TV allocations
Table 16.4 shows a simplified representation of present European UHF two-way radio and radio/TV allocations.
Table 16.4 European UHF two-way radio and radio TV allocations
Part of the debate is now focused on whether LTE can deliver the required functionality for this specialist market and somehow combine that with global-scale economy. This in turn requires some commonality to be achieved between US, European, Asian and rest of the world band allocations.
In the US, discussion has focused on whether or how to deploy LTE for a public-safety national wireless broadband network and regional networks in the 700-MHz band. The US debate provides a reference for the discussions that are taking place on LTE public-safety networks in Europe, Asia and the rest of the world and provides an insight into some of the spectral and standards issues that need to be resolved in order to translate present market ambition into practical reality. As can be seen from Tables 16.5 and 16.6 there is no commonality between the US band plan and the LTE 700 band plan in Asia.
Table 16.5 Upper band 700-MHz LTE including the public-safety bands
Table 16.6 LTE 700 band in Asia (Region 3)
16.12 The LTE 700 MHz Public-Safety-Band Plan
It has been just over ten years since the 9/11 attacks in the US. The attacks prompted a major reassessment of public-safety communication needs that have been at least partially reflected in spectral allocation and auction policy and to a lesser extent in technology policy.
The upper 700-MHz band plan adopted by the FCC on 1 July 2007 allocated a 2 by 5 MHz duplex band at 763 to 768 and 793 to 798 MHz for a single nationwide public-safety broadband network and a second duplex band known as D band/Band 14 at 758 to 763 and 788 to 793 MHz to be auctioned with public-safety requirements. Both channel pairs work as reverse duplex with mobile transmit in the upper duplex. This means that the two lower channel pairs supporting base transmit (the downlink) are adjacent to Verizon's ten MHz of C band LTE 700 downlink (Band 13).
There are also proximate 6-MHz allocations for regional and national narrowband networks, dangerously proximate in terms of coexistence between Verizon upper band LTE transmit and the narrowband public-safety receive channels. These narrowband allocations are technology neutral, but theoretically could support either 2 by 3 MHz or 4 by 1.4 MHz LTE channels, not exactly narrowband but relatively narrowband when compared to other LTE networks.
The US LTE 700-MHz band plan between 698 and 806 MHz is unlikely to be adopted in Asia (with very good reason). The most likely band plan in these markets (basically most Asian markets excluding Japan) is a 2 by (rather ambitious) 45 MHz duplex pair implemented as a standard duplex as shown in Table 16.6.
Putting this US regulatory setback to one side, the promise of LTE public safety spectrum in the US market and associated federal funding has prompted infrastructure vendor interest.
In July 2010, Motorola announced a pilot LTE public-safety network in San Francisco as a broadband overlay to the existing APCO 25/Project 25 narrowband digital radio system. APCO25 is a user requirement body overseen by the Association of Public Safety Communications Officials. Project 25 (P25)1 fulfils a similar role to the TETRA Forum2 and ETSI TETRA standards process3 in European and ROW markets. P25 radios have overlapping spectrum with analogue and iDEn digital two-way radios. The networks and band allocations are also described functionally as public protection and disaster relief (PPDR).
At the APCO Conference in August 2010 there were vendor announcements from Alcatel on public-safety LTE integration with EADS supply public-safety 911 despatch data systems. EADS also supplies TETRA systems outside the US. Nokia Siemens announced joint projects with Harris, a P25 supplier. On 7 September Motorola and Ericsson announced they would be working together on LTE public-safety solutions
Public-safety vendors and regulators have a chequered history when it comes to delivering interoperability. TETRA and P25 radios for example are incompatible both in terms of technology (the air interface) and band allocation. TETRAPOL does not interwork with TETRA. Within the US and Europe and Asia, legacy analogue networks are still operational. Interoperability can often therefore only be realised by supporting multiple digital and analogue radio systems making scale economy hard to achieve.
In the US, the white-space initiative (see Chapter 15) sometimes also described as super WiFi will introduce complexity and uncertainty when managing multiple radio system coexistence particularly if public-safety networks are designated as safety critical.
The LTE 700 MHz public-safety networks could work completely separately from all other radio systems but from a specialist user perspective it would be sensible to have at least some interoperability with other networks. For example, public-safety rural data coverage requirements and user expectations would suggest additional infrastructure will be needed in rural areas. Conversely, public-safety bodies are unlikely to have the finance to build dense urban networks. Public safety users will expect and probably insist on broadband connectivity with the same coverage footprint as existing narrowband voice networks. Economically, this means that public-safety LTE 700 networks and commercial cellular LTE 700 networks will need to find some way of working together technically and commercially.
Table 16.5 suggests the obvious starting point for the US market at least would be a dualband radio capable of accessing Verizon LTE upper C band (Band 13). However, this would increase the operational pass bandwidth of the Verizon Band 13 radios by 18 MHz. This will compromise performance and add cost and weight, an unattractive option for Verizon and Verizon subscribers.
16.13 The US 800-MHz Public-Safety-Band Plan
An alternative option is to produce dualband P 25/LTE or iDEN/LTE radios.
In terms of band allocation, legacy APCO 25/P25 radios are implemented with mobile TX at 821 to 824 MHz (immediately adjacent to the US 850 cellular mobile transmit band between 824 and 849 MHz) and mobile RX at 866 to 869 MHz (immediately adjacent to the US 850 mobile receive at 869 to 894 MHz).
However, Release 9 of the 3GPP specifications proposes three extensions/variations to the US 850 band including adding 10 MHz at the lower end of the present Band 5 allocation that includes all of the legacy P25 800 MHz channels and some of the proposed new channels. P25 radios and iDEN radios are now coming to market capable of supporting the frequency bands shown in Table 16.7. This extends P25 and iDEN into channels previously used and still used for analogue two-way radio. The table also shows the impact of the proposed extension of the US 850 cellular band.
As stated earlier iDEN radios have 25-kHz channel spacing with a transmit power of up to 1 watt and a six-slot QPSK modulated TDMA physical layer.
Table 16.7 US 800-MHz public safety – old and new band plan
P25 radios can have 12.5, 20 or 25 kHz channel spacing and an output power of between one and three watts (also QPSK modulated). In-car mobiles can have an output power anywhere between 10 and 35 watts. Base stations can be up to 100 watts. This is serious radio, optimised for rural coverage and deep urban inbuilding penetration.
Multiband versions of P25 radios also have to be capable of working at VHF between 136 and 174 MHz and UHF between 380 and 470 MHz. The theory is that LTE will provide a broadband bridge between these legacy networks and will allow the public-safety sector to benefit from global-scale economies with an opportunity to amortise research, development and manufacturing investment across billions of user devices per year. From a specialist user-experience perspective this should provide an opportunity to source standard form factor phones, smart phones, tablet/slates and lap tops at consumer prices.
Ideally, this user equipment would also be able to access commercial LTE networks.
It would seem initially attractive to consider integration with the US 850 band rather than LTE 700. Future iterations of P25 could include a 5-MHz LTE channel adjacent to the US 850 commercial networks as or when or if they transition to LTE.
The problem with this from an RF performance perspective is that even extending the US 850 band by 10 MHz to 35 MHz decreases user equipment sensitivity by 1 dB. Here, we are suggesting adding 18 MHz to create a passband of 43 MHz. This would involve at least as much loss again.
Coupling P25 radios with the LTE channel allocations in the 700-MHz band would require a dual-mode radio to support P25 TX at 806 to 824 MHz and theoretically at least low-band LTE 700 TX down at 698 MHz. This is 126 MHz of operational bandwidth equivalent to a 16% ratio of the centre of the band. The stretch on the RX path would be similar. This is technically challenging and commercially unlikely. In practice, the user equipment would be several radios in one box.
A P25 dual-mode radio that interoperated just with the LTE national public safety 700-MHz network would be easier to implement and the narrowband public-safety bandwidth could be included as well, but the devices would still need to stretch on the receive path from 758 MHz to 869 MHz (111 MHz) or 746 MHz to include Verizon Band 13 (123 MHz). Either option would still need a dual or triple front end to be sufficiently efficient.
The problem could be solved technically by throwing away the P25 and iDEN radios, reallocating the bandwidth to US 850 operators and just having LTE 700 police and SMR radios. However, this would be hard to sell to public-safety and specialist mobile radio users. They would quite rightly point to some of the difficulties encountered when Sprint Nextel iDEN users were forcefully migrated on to the Sprint CDMA network.
Alternatively, the LTE mobile broadband network could be regarded as a separate network function with legacy networks and new P25 and iDEN radios providing voice coverage and specialist user requirements such as all informed user capability, <250 millisecond press to connect onto channel access times, multigroup and wide-area broadcast messages, dynamically changeable priority levels, extra secure authentication and encryption, task-optimised noise cancellation, storm plans and special-event plans, back-to-back working, ground-to-air links, surveillance functions and ruggedised user equipment – all that special stuff. This, however, implies parallel standards and parallel development, manufacturing and test processes4 and significant interworking and interoperability challenges.
An alternative would be to use LTE for broadband and narrowband, including voice but voice coverage both in rural areas and in terms of urban inbuilding penetration would need to be absolutely as good as existing narrowband specialist radio.
LTE achieves high peak data rates by using wide channels (up to 20 MHz) in a wide variety of widebands (190 MHz at 2.6 GHz for example) with longer-term support of eight or more bands. The ‘cost’ of delivering these high peak data rates is that multiband LTE user equipment will be inherently less sensitive than single-band narrowband radios.
Specialist mobile radio networks have to support fewer bands, often in the past only one, and have much narrower operational bandwidths and channel spacing. Narrow channel spacing allows for a narrowband IF. A narrow operational band allows for high-Q filters and a highly efficient transmit and receive chain. This means that sensitivity and selectivity and transmit and receive efficiency are generally better than wider-band devices. As a result, voice coverage in specialist radio networks is likely to be better, particularly in rural areas or for deep inbuilding penetration.
The IP voice used in LTE networks introduces additional packet header overheads that have to be taken out with a compression algorithm. Extracting a 12-kbps voice stream from a wideband 20-MHz channel is also computationally expensive. The computational overheads and compression clock cycles will also introduce delay overhead that would be better avoided if press to talk/press to connect needs to be supported. Talk groups and back-to-back working will also be needed, which will require additional standards work.
For applications where high peak data rates are needed, LTE will provide peak data rates that specialist radio networks will find it impossible or at least difficult to match.
For voice and text applications, or where average data rates are more important than peak data rates, for example in larger cells, narrowband networks will provide better voice quality and coverage and probably a longer data duty cycle. LTE could compete on coverage and building penetration if the voice or data traffic was heavily error protected, but this will compromise spectral efficiency. If the bandwidth is inexpensive this does not matter. If the bandwidth is expensive, it matters a lot. Heavy error protection will also reduce TX and RX efficiency, which will compromise the user data duty cycle. This does not matter if the user is attached to a vehicle power supply. It does matter if the user is out on an eight-hour shift on foot or bike or horseback (not an uncommon requirement in crowd control situations).
16.14 Policy Issues and Technology Economics
So the suitability of LTE for public safety in the US depends partly on spectral allocation and auction policy, partly on technology policy, partly on the application mix and partly on available budgets.
It seems unlikely that LTE 700-MHz networks on their own could meet all the needs of the public safety and/or SMR sector. This means that it will be impossible to discontinue either P 25 or iDEN radio and network system development and deployment. This loads parallel product development manufacturing and test cost onto the sector.
Some of this cost could be reduced if future iterations of P25 and iDEN could be dovetailed into the Release 10 (LTE Advanced) standards process. A 5 MHz channel could be used for LTE leaving sufficient channel bandwidth for narrowband Project 25 voice connectivity.
This would of course be entirely possible but presently improbable due to understandable vested interest both in the specialist radio vendor and specialist user community. Sector procurement policy is also understandably cautious about what might be perceived as relatively radical change.
While it is possible to design military specification radios that can access multiple bands using multiple technologies, these devices are neither low cost nor particularly energy efficient. They are also heavy. Public safety and emergency-response user requirements could be best met by flexible radios capable of accessing commercial LTE radio bands and the public-safety bands with RF performance at least as good as existing specialist radio products at a price point close to consumer price expectations. At present it seems unclear as to how this will be achieved.
The US 700-MHz band plan is a salutary example of what happens when spectral allocation and auction policy is decided by economists who ignore engineering advice and abrogate regulatory responsibility by allowing the market to decide technology policy. Out of necessity, operators and vendors have to respond to short-term market expectations. These are incompatible with the long-term decisions that need to be taken on future technology options.
The US may well end up with well-executed and well-integrated broadband and narrowband public-safety networks but they will be neither technically nor commercially efficient unless a much greater degree of user equipment RF front-end flexibility can be realised. This in turn requires a level of investment to which the industry presently has limited visibility.
In Europe and the Rest of the World there is an obvious opportunity to use at least part of the LTE 800 band (between 790 and 862 MHz) to support public-safety broadband connectivity particularly as these bands overlap the Project 25/iDEN bands in the US.
Table 16.8 US and European/ROW 800-MHz band plan implemented as standard duplex
| US | EUROPE/ROW | |
| MOB TX P25/iDEN/US850 | MOB TX LTE 800 | |
| 790 | ||
| 806 to 809 | Narrowband P25 or iDEN
Could also be two by 1.4 or one by 3 MHz LTE |
|
| 809 to 814 | Broadband P25 LTE or iDEN LTE (5 MHz) | |
| 814 to 849 | Extended US 850 LTE | 821 |
| MOB RX LTE 800 | ||
| 831 | ||
| 851 to 854 | Narrowband P25 or iDEN
Could also be two by 1.4 or one by 3 MHz LTE |
|
| 854 to 859 | Broadband P25 LTE or iDEN LTE (5 MHz) | |
| 859 to 894 | Extended US 850 LTE | 862 |
| MOB TX LTE 900 | ||
| 880 | ||
| 896 to 902 | iDEN SMR TX | |
| 915 | ||
| MOB RX LTE 900 | ||
| 925 | ||
| 934 to 940 | iDEN SMR RX | |
| 960 |
This would mean that standard LTE 800 user equipment and network hardware and software could be used in the US public-safety market in the APCO band, providing the global-scale economy needed by the sector.
The (rather major) snag with this is that LTE 800 may be deployed as a reverse duplex with mobile RX at 790 to 821 MHz and mobile TX at 831 to 862 MHz. This is because there is concern in the broadcasting community that transmissions from LTE mobile users will interfere with terrestrial TV reception, a process known as ‘hole punching’. Alternatively, the band could be deployed as a standard duplex with TV signals retransmitted from cellular base stations. Apart from opening up opportunities for integrated LTE 800/Project 25 public-safety networks we would also get reliable digital TV on portable receivers – always handy in a national emergency and particularly relevant in countries presently attempting to reinvigorate local TV. It would also make TV more resistant to interference from white-space transmissions. The band plan if implemented with LTE 800 as a standard duplex (mobile TX in the lower paired band) would look something like Table 16.8.
As an added bonus it can be seen that LTE 900 overlaps with the upper iDEN bands at 900 MHz, which could be implemented as four by 1.4 MHz LTE bands, two by 3 MHz LTE or one by 5 MHZ LTE with some guard band.
Getting a band plan like this underway would, however, require a regulatory environment that creates incentives to encourage cooperation between the cellular and broadcasting industry, the public safety radio user, standards and vendor community and white-space vendors and investors on both sides of the Atlantic, preferably bridging private-sector and public-sector interest.
An adversarial spectral allocation process designed to maximise income from spectral auctions makes this cooperation harder to achieve. This is compounded by a failure to realise any substantive global harmonisation, particularly in the LTE 700 band. A lack of cross-sector and/or transatlantic let alone global thinking and direction in standards setting also does not help.
International spectral and standards policy together have a direct impact on user equipment cost and performance and user experience value. Regional standards now make very little economic sense. Nationally specific standards make even less sense. Vendor-specific standards only make sense to the vendor.
If a local or regional band allocation or adopted standard does not scale on a global basis, this needs to be factored into bid valuation and network return on investment expectations. This seldom seems to happen.
16.15 Satellites for Emergency-Service Provision
On 10 July 1962 Telstar was launched into an elliptical orbit going round the earth every two hours and 37 minutes at a height of between 1000 and 6000 kilometres. The satellite had been built at Bell Telephone Laboratories, was roughly spherical, about 34 inches (880 mm) in length and weighed 170 pounds (77 kilograms). Working on 14 watts of power from a solar cell array the satellite received signals at 6 GHz and transmitted at 4 GHz and successfully relayed television pictures, telephone calls and faxes.
The launch was privately sponsored, which in itself was an innovation. The consortium included AT and T, the British Post Office and the French National PTT. The US ground station was in Andover in Maine, the UK ground station was on Goonhilly Downs and the French Ground Station was at Pleumeur-Bodou in North Western France. The BBC managed the conversion between the 525 line (US) and 405 line (UK and European) television standards. The ground-station antennas were huge, weighing 380 tons, with an aperture of 3600 square feet (330 square metres). The antennas were housed in radomes the size of a 14-storey building and had to track the satellite with a pointing error of less than 0.06 degrees.
Sixty years later satellites continue to play a fundamental role in telecommunications, but probably most impressively are powerful enough to transmit signals to hand-held devices. It is this part of the story that we particularly want to tell.
Just over thirteen years ago two companies, Iridium and Globalstar started providing a service to mobile users with hand-held phones from two low-earth orbit constellations.
Figure 16.1 The Iridium Constellation. Reproduced with the permission of Iridium.
The Iridium5 project, championed, engineered and financed by Motorola, involved launching sixty six satellites into low-earth orbit to provide cellular-type services at a time when cellular networks were becoming increasingly ubiquitous and cellular service increasingly competitively priced. The system was and still is a spectacular engineering success; a tribute to largely US-based engineering resource, but at the time was a fiscal failure. Figure 16.1 shows the constellation and north south orbital paths.
The business model was predicated on the existence of a user community who would prefer not to use their cellular phone or two-way or short-wave radio as a preferred communications device. This user community would instead choose a system where the phones and phone service were made available at a substantial premium with poor indoor coverage, packaged in a form factor similar to a Motorola World War walkie-talkie radio.
Globalstar launched a competing constellation of 48 higher-altitude satellites. Like Iridium these were an engineering triumph but at the time a fiscal failure. Both Iridium and Globalstar went into Chapter 12 administration.
But life moves on and moves in mysterious ways. A retrospectively prescient decision was taken not to deorbit the satellites but to maintain both constellations and continue to service and develop a loyal group of specialist users. And then came 9/11, and the second Gulf War and Afghanistan and Hurricanes Katrina and Rita and the Asian Tsunami, the Madrid bombings and the 7/7 bus and tube bombings in London and more recently the forest fires in California plus earthquakes in Turkey, Haiti, China, New Zealand and Japan, a few famines and floods and other natural and unnatural disasters around the world.
16.16 Satellites and Cellular Networks
Cellular networks were, and are, not always ideal to provide first-responder support in these often hazardous and naturally or unnaturally chaotic unwanted and unpredictable events. For example, towers and/or terrestrial telephone links can be blown up or blown down.
Iridium has always had a number of inherent resiliency advantages both over terrestrial only networks and other satellite networks. It was, and is, the first and only civilian low-earth orbit satellite system to implement intersatellite switching, reducing dependency on any single ground facility.
As a LEO (low-earth orbit) constellation), round-trip latency is 20 milliseconds, substantially lower than the 133 milliseconds of a MEO (medium earth) or the 500 milliseconds of a GEO (geostationary) satellite system. This makes speech and latency-sensitive data exchanges easier to support.
Additionally, the satellites have proved to be significantly more robust than expected and continue to provide service across the US, Alaska, Hawaii, the Pacific Ocean (as an integral part of the now updated tsunami warning system) and other hard-to-reach parts of the world.
Iridium therefore had a perhaps unexpected opportunity both politically and financially to justify new investment in a replacement constellation and updated service platforms, to negotiate innovative collaborative deals with other traditional and nontraditional service providers and possibly in the longer term to justify preferential access to new spectral allocations at L band between 1518 and 1675 MHz or S band between 1.97 and 2.69 GHz. Iridium and a number of other entities have similar plans. Iridium have executed on that opportunity with a fully funded proposal to launch a new constellation.
Whatever we write at this point will be out of date by the time you read it but essentially the new constellation (called NEXT) will do everything the old constellation did but better and with enhanced data and environmental sensing functionality. The best thing is to go to the Iridium NEXT6 web site to see what has happened.
16.17 The Impact of Changing Technology and a Changed and Changing Economic and Regulatory Climate – Common Interest Opportunities
This opportunity has to be seen within the context of a substantially changed and changing economic and regulatory climate. Satellites are attractive again as investment opportunities.
Partly, this shift is technologically driven.
Satellites can now pack more processing power into a much smaller space. Advances in RF and baseband hardware have delivered a steady year-on-year increase in functionality per kilogram of orbital weight. Solar panel arrays are more efficient and can deliver more onboard power to support wider-bandwidth two-way communication.
Smart antenna technologies have improved over the past ten years so available power can be more accurately and adaptively deployed. Improvements in station-keeping efficiency and hardware reliability have helped to increase the life span of satellites. An operational life of 15 years is now a realistic expectation even for the traditionally shorter-lived low-earth orbit platforms.
A reasonably broad choice of launch options and some innovative mission-insurance solutions have helped trim launch costs. All these factors together have contributed positively to the overall economics of providing or updating and upgrading satellite-based services.
Iridium has the advantage of an existing constellation, an established and loyal user base and a track record of providing emergency-service support.
It has to be said that cellular operators have not been as conspicuously successful at nurturing and serving specialist user communities. The lack of service immediately post Hurricane Katrina for example was understandable but resulted in politically costly censure.
Cellular operators would do well to review their service offerings for the public-safety sector and ensure that these sectors are at least adequately represented in their overall customer mix. Just focusing on consumer and corporate users and/or consumer and corporate applications is probably not wise either financially or politically in the present unstable world climate. This suggests an opportunity for cellular operators to work with satellite service providers on integrated services for specialist users.
Conversely, Iridium have to manage their user base across a transitional period where the existing constellations are past their nominal end-of-life expectancy. Globalstar, a competitor constellation had a number of RF hardware failures that reduced availability of their S band voice services and both Globalstar and Iridium have to finance and launch new satellites over the next five years. At the time of writing Iridium appear to be successfully achieving this.
This suggests a need to work with rather than against cellular operators. Each party has something the other party needs, always a good basis for a collaborative venture. There may be additional opportunities to work with other satellite operators with medium-earth or geostationary satellite systems, two-way radio service providers and the broadcasting community. Most broadcasters for example have a public-service remit that extends to providing emergency broadcasting services in response to local, national or international emergencies.
Satellites have always been a politically sensitive sector and so has satellite spectrum, particularly the allocations in L band and S band which are shown in Table 16.9. Some of that spectrum has been acquired or allocated on advantageous terms and can potentially be repurposed beyond an original remit to provide specialist broadcasting and/or emergency-service provision. For example, spectrum originally intended for broadcast TV could be extended to embrace a much broader multiplex of essential and nonessential service propositions.
Table 16.9 Terrestrial broadcast and satellite coexistence in L Band
There is substantial scope here for special pleading on the basis of social need, for example the provisioning of broadcasting and/or emergency service communications in emerging countries. Special pleading can, however, sometimes be a smoke screen for more prosaic economic ambition.
Cellular operators are right to be wary of new competitors who could be potentially successful at leveraging political influence into preferred access to new or existing spectrum. To an extent the very specific public service obligations imposed on the public-safety bands in the upper band UHF US auction are an early indication of similar battles that will be fought internationally over the next three to five years in the ongoing L band and S band allocation and auction process.
Such an adversarial approach to spectral allocation is inappropriate in a world where radio communications, particularly integrated radio communications, have an increasingly important role to play in emergency-service provision.
16.18 And Finally – Satellite and Terrestrial Hybrid Networks
Here, unashamedly, I returned to a 2007 technology topic on satellite and terrestrial hybrid networks, a sector that I have always been sceptical about.
In 2007 this was a gung ho investment opportunity. One of the big players who emerged in the 2007 to 2011 ‘window of opportunity’ is a company called LightSquared who inked an agreement with Inmarsat.
Table 16.10 GPS satellite constellations
I cannot resist including this particular news item by Ian Scales dated 21 June 2011 from Telecom TV.com. Table 16.10 puts the GPS issue into perspective.
LightSquared management and its investors are either on a roller-coaster ride of “ducking, diving, dodging and weaving,” as those with South London entrepreneurial spirit might describe it, or they are following a master plan of great cunning and complexity.
In case you've not been following, LightSquared is a US start-up, backed by a prominent investor fund, which has developed plans for the building of a US-wide wholesale ‘hybrid’ LTE network, using both conventional LTE deployment (base stations, femtos, etc.) and satellite-delivered LTE for those hard-to-get-at places. With this carrot (rural coverage) it has secured regulatory clearance, investor and vendor backing and, apparently, contingent AND concrete wholesale customers (operators who will buy capacity from LightSquared to sell on to their customers). LightSquared also seemed to move quickly from PowerPoint to BuildPoint, with a satellite launched and deals with other carriers and equipment vendors signed.
But the ‘Radio Network God’ (whoever she is) is not easily placated by a slick presentation and few partner announcements. LightSquared ran into trouble over the interference problems it would allegedly visit on the Global Positioning System which operates in adjacent spectrum.
A campaign to stop LightSquared was promptly mounted which turned into a ginger group (SaveOurGPS.org, described as a cross-industry collection of manufacturers) and the company agreed to independent testing to make sure there was no interference.
Recently, it was supposed to report back to the FCC on the results, but has been delayed on the grounds that LightSquared needed to consider and analyse, leading to obvious speculation that the satellite part of the plan had been gotcha’d, interference was indeed a real problem, and that someone, somewhere had some explaining to do to LightSquared investors on why the technical due diligence failed.
To bolster confidence, details were then leaked on a complex network sharing deal with Sprint, which seemed to indicate that the terrestrial part of the enterprise was still on track and viable, no matter what happened with the satellite part (which will now, no doubt, be presented as a peripheral aspect of the build plan).
A cynical observer might conclude that leaked details are often preferable to properly presented details because crucial flaws can remain obscured behind a wall of “no comment, we are in a quiet period” (a stance that has become more company moto than PR tactic, and deployed by many companies, not just LightSquared).
Now, the very latest twist in tail circuit is that LightSquared has a produced satellite spectrum ‘Plan B’ that nobody appears to have guessed existed. The line is that instead of using the spectrum it was initially going to use for the satellite part of its LTE service, it has produced another chunk of spectrum which will serve almost as well and is parked further away from the spectrum used by the Global Positioning System so won't interfere with it. This spectrum was to be used later on in the system's development plan, the company has indicated.
To bring this 10-MHz block of spectrum into operation, however, it will have to work with satellite operator, Inmarsat, to roll out services.
So problem solved? Of course not.
SaveourGPS.org says the new plan is a figleaf. Even at the lower end of the MSS band (which it's using) the LightSquared service would still interfere with many GPS receivers as well as the precision receivers that LightSquared itself has conceded will be affected. The results of the interference study, it says, already confirm this – the only way forward is for LightSquared is to move out of the MSS band completely.
This story has some way to run yet.
The moral of this tale is of course it is really important to do technology due diligence before you throw millions and sometimes billions of dollars at a business plan.
But now let's go back to that July 2007 technology topic.
In July 2007 dual-mode networks that combine satellite and GSM cellular services already existed. Thuraya, a network operator servicing the United Arab Emirates and ACEs servicing Asia were two examples.
Hybrid satellite/terrestrial systems are different in that they use terrestrial repeaters to combine the wide-area coverage capabilities of satellites with the urban coverage and in building capabilities provided from terrestrial networks. These may or may not be associated with existing cellular or terrestrial broadcast networks.
The terrestrial repeaters are described in the US as ancillary terrestrial components (ATC) and in Europe and Asia as ground-based components. In China the networks are described as satellite and terrestrial interactive multiservice infrastructure (STiMI).
Table 16.11 Present and proposed S band satellite and terrestrial (ATC) broadcast and two-way radio networks
Hybrid networks were already used in the US to deliver audio broadcasting for in-car and in-flight entertainment. Widespread deployments of these systems were planned both for audio and TV broadcasting and for two way cellular service provision.
These deployments exploited already allocated L Band and S Band spectrum in US and rest of the world markets. Some of this spectrum has been gifted to the operators. Typically these allocations are not owned by traditional terrestrial cellular or terrestrial broadcast service providers. As such, they represented a competitive threat and by implication a collaborative opportunity for the cellular network operator and traditional terrestrial broadcast community. Present S band satellite spectral allocations are shown in Table 16.11. Their relationship to cellular allocations above and below 2 GHZ are shown in Table 16.12.
Table 16.12 US S Band cellular allocations including satellite/terrestrial hybrids
| Band | Frequency (MHz) | Operational Bandwidth (MHz) |
| AWS | 1710–1755 | 45 |
| PCS 1900 | 1850–1910 | 60 |
| Sprint/ Nextel | 1910–1915 | 5 |
| PCS1900 | 1930–1990 | 60 |
| Sprint/ Nextel | 1990–1995 | 5 |
| ICO/Terrestar | 2000–2020 | 20 |
| AWS | 2110–2155 | 45 |
| ICO/Terrestar | 2180–2200 | 20 |
| WCS(Sprint/Nextel) | 2305–2320 | 15 |
| SDARS (XM/Sirius) | 2332.5–2345 | 12.5 |
| WCS (Sprint/Nextel) | 2345–2360 | 15 |
| WiFi | 2400–2480 | 80 |
Hybrid networks were, and possibly are, an integral part of the convergence presently taking place between satellite and terrestrial cellular, TV broadcast and broadband and narrowband data-delivery systems. This convergence offers positive crossover value opportunities but these opportunities need to be qualified within the present and future context of the satellite industry. They also need to be technically efficient. If a network is not technically efficient it will not be commercially efficient.
16.19 Satellite Spectrum and Orbit Options
The satellite systems of interest to us are deployed either into L Band between 1518 and 1675 MHz or S Band between 1.97 and 2.69 GHz and are either in geostationary geosynchronous (GSO) orbits at 35 000 km, in medium-earth (MEO) orbits between 10 000 and 20 000 km or in low-earth (LEO) orbits or ‘high’ low-earth orbits (HLEO) between 700 and 1400 km. For the sake of comparison, the US Space Shuttle orbits at 350 km.
Some highly elliptical orbits such as the Molniya or low polar orbits provide optimised coverage for countries at extremely northern or extremely southern latitudes. ‘Seeing’ geostationary satellites from these latitudes can be problematic. The choice of orbit determines the number of satellites needed to provide a particular coverage footprint, the size and position of the satellites determines their functionality. The spectrum into which the satellites are deployed and the proximity of this spectrum to other users also determines their functionality, particularly in terms of system interoperability with other terrestrial or satellite networks.
LEOS have the advantage of low round-trip latency, about 20 milliseconds compared with 133 milliseconds for a MEO and 500 milliseconds for a GSO satellite. Geostationary satellites have the advantage that they stay in the same place when viewed from the earth. This simplifies handover and radio planning algorithms when servicing mobile users.
The size of the satellite determines the size of the antenna array, sometimes upwards of 20 metres in large geostationary satellites. The size of the array determines the uplink and downlink gain available, particularly if adaptive spot beam-forming techniques are used.
Similarly the size of the solar panel array, sometimes upwards of 40 metres in large geostationary satellites, dictates the amount of power that can be generated, which determines both the coverage and capacity.
Advances in launch technologies have made it possible to launch satellites weighing over 5000 kg into geostationary orbit. Advances in RF and baseband hardware deliver a steady year-on-year increase in functionality per kg of orbital weight.
Conversely, microminiaturisation techniques have made possible new generations of super small satellites though these tend to be used for more specialist low-orbit or deep-space exploration applications. However, terrestrial network hardware costs have also reduced over the past twenty years by roughly 15% year-by-year and functionality has greatly increased.
In particular the rapid growth in subscriber numbers served by terrestrial cellular networks and terrestrial broadcast networks has attracted engineering investment that in turn has improved the delivery cost efficiency of these networks. Present upgrading of the DAB terrestrial networks in the UK to provide improved coverage and higher data rates provides an example.
In terms of user devices, the economies of scale available to the cellular industry effectively dictate the radio functionality included in mobile handsets. In the past, these factors have invalidated a number of apparently promising satellite based business models. New satellite ventures therefore have to be approached with caution and the relative merits and demerits of each option need to be carefully considered. This is particularly true when significant amounts of spectrum are being allocated by regulators either for satellite-based services or for new hybrid satellite terrestrial network propositions.
Satellites have delivered telecommunications, TV and data for over 50 years, the original triple-play proposition. This interdependency has determined the economics of the industry.
Over the past twenty years improved power output (downlink capability) and sensitivity (uplink capability) has allowed satellites to play an increasing role in delivering communications to mobile devices including mobile handheld devices. Hybrid satellite terrestrial networks are a logical next step, but have to be qualified in terms of the additional value that they deliver to existing terrestrial networks.
16.20 Terrestrial Broadcast and Satellite Coexistence in L Band
Table 16.9 referred to earlier shows the allocations to terrestrial broadcast, satellite broadcast and satellite two-way services in L Band.
16.21 Terrestrial DAB Satellite DAB and DVB H
The L Band allocations for DAB and DVB H were identified, although these allocations were not universally available in all countries. DAB was intended primarily for radio but could carry a TV multiplex of up to 7 video channels on a 1.7-MHz OFDM channel.
DVB H was implemented as a trial in New York delivering 75 TV and music stations transmitted from 74 terrestrial sites covering 475 square miles. The network used a 5-MHz channel rather than the standard 8 MHz used for DVB H at UHF or S band. DVB H could have been implemented in L band as a DVB SH network with terrestrial repeaters but lacked global scale and therefore could not be progressed.
16.22 World Space Satellite Broadcast L Band GSO Plus Proposed ATC
World Space provided radio services to Africa and Asia from two geostationary satellites, Afristar launched in 1998 and Asia star launched in 2000. These satellites use the same air interface as XM radio. Plans to launch a third satellite, Ameristar, to serve South America were not implemented as these L band frequencies were used by the US Air Force.
This satellite was then intended to be repurposed to provide European coverage with a particular emphasis on Italy. The satellite would transmit and receive with an air interface theoretically compliant with the ETSI satellite digital radio standard that was being extended to include S UMTS and DVB SH and a legacy set of standard documents known as GMR, the geostationary mobile radio standard. World Space had terrestrial repeater licenses for Bahrain and the United Arab Emirates and lobbied for similar license concessions in its other addressable L band markets.
16.23 Inmarsat – L Band GSO Two-Way Mobile Communications
Inmarsat have traditionally provided mobile and fixed communication services to the maritime, aeronautical, land mobile and remote-area markets. Recent investments have focused on increasing data rates as part of their Broadband Global Access (B GAN) network.
Space to earth links are at 1525 to 1559 MHz and earth-to-space are at 1626.5 to 1660.5 MHz. The network supports user data rates up to 256 kbps. Terminals are typically 1.3 kg or above and power consumption is in the region of 14 watts. There are ten satellites in a geostationary (GSO) orbit. The latest satellites launched have a planned 13-year operational life and a 9 metre antenna array.
16.24 Thuraya 2 L Band GSO Plus Triband GSM and GPS
Thuraya provided satellite service to mobile hand-held devices that also supported triband GSM and GPS with coverage optimised for the United Arab Emirates and the Middle East. The original geostationary satellite was launched in 2000 with a second in 2003 and a third in 2004. Uplinks and downlinks were the same as Inmarsat. The Thoraya 2 satellite had/has a 12-metre antenna array. The GPS frequency at 1575.42 sits reasonably conveniently between the L band uplink and downlink frequencies used in the handset. There were no stated plans for a hybrid ATC network.
16.25 ACeS L Band GSO Plus Triband GSM and GPS
More or less equivalent to Thuraya combining service from one geostationary satellite launched in 2000 and positioned to provide optimal coverage over Asia rather than the Middle East. Inmarsat and Aces proposed to combine their network proposition to provide global coverage. There were no stated plans for a hybrid ATC network.
16.26 Mobile Satellite Ventures L Band GSO Plus ATC
Mobile Satellite Ventures planned to launch two geostationary satellites in 2009/2010. These were large satellites weighing 5500 kg with a 22-metre antenna array and 500 spot beams. Coverage was to be optimised for the US, Southern Canada and Latin America.
The company had a license to deploy a 30-MHz paired band with the lower duplex between 1525 and 1559 paired with the upper duplex between 1625.5 and 1660.6 MHz based on proposals first tabled to the FCC in 2001. This became the basis for the LightSquared business case.
The mobile uplink was to be processed by both satellites with the signal being diversity combined, a technique also used by Globalstar. Terrestrial transmitters were to be added to existing cellular sites to provide additional coverage and capacity. Essentially this was an extension of the terrestrial repeater principle to two-way communication based on a combination of terrestrial cells with a radius of between one and five kilometres nesting within a satellite cell with a radius of about 100 km. This brought Mobile Satellite Ventures into direct competition not only with terrestrial broadcasters but also with cellular service providers. Mobile Satellite Ventures claimed to have substantial patent-based intellectual property regarding space/terrestrial frequency reuse, beam forming with larger arrays and handover algorithms. There was some crossover of patent value and personnel with XM radio.
16.27 Global Positioning MEOS at L Band GPS, Galileo and Glonass
Table 16.10 referenced earlier summarized the global positioning MEOs.
The US-managed GPS satellites have a 7.5-year design life, but satellites are lasting 12 years.
Galileo is a European initiative with coverage, positioning accuracy and satellite lock times optimised for Europe. The satellites will have a 12- to 15-year design life. Galileo has the same downlink frequency as GPS and could be expected to be implemented as standard in future handset designs.
It is theoretically convenient to build an L band transceiver with GPS capability, though care has to be taken to avoid desensitisation of the GPS or Galileo signal within the handset. This appears at the time of writing to be the fatal flaw in the LightSquared business plan.
GPS is of course now in a wide range of devices including the latest Sony ‘Vita’ PlayStation (Chapter 9). Most smart phones and of course all navigation devices, usually integrated with a GPRS receiver to capture traffic jam alerts to support optimised routing. It seems like we may be the last generation to be able to read a map.
16.28 Terrestrial Broadcast and Satellite Coexistence in S Band
Table 16.11 summarised present and proposed S band hybrid satellite and terrestrial broadcast and two-way radio networks
16.29 XM and Sirius in the US – S Band GEO Plus S Band ATC
XM7 and Sirius8 are two operators in the US providing MP3-quality audio radio for in-car and more recently in-flight entertainment. The two companies are presently in merger discussions.
XM has four 15-kW geostationary satellites XM1 (Rock), XM2 (Roll), XM3 Rhythm and XM4 ‘Blues’. XM1 and XM2 have suffered some fogging on their solar panels.
The satellites work with 1500 terrestrial repeaters that are each technically capable of delivering an ERP of 25 kW. Sirius has an additional three satellites also in geostationary orbit.
The air interface is specified in a standard known as SDARS9 (Satellite Digital Audio Radio Service). The networks operate in a 12.5-MHz band between 2332.5 and 2345 MHz with four of the six radio carriers dedicated to the satellites and the other two channels dedicated to the terrestrial repeaters.
Satellites are QPSK to maximise power efficiency. The terrestrial repeaters use COFDM.
User terminals have two separate antennas, one for the satellite signals and one for terrestrial signals. The signals are combined in the receiver at baseband using maximal ratio combining. Alternatively, simple voting is used to choose the strongest signal.
The receivers tend to be traditional superhets with a 75-MHz IF commonly also used in TV receivers. Antenna systems for these devices are complex10 though the design requirements for receiving satellite and terrestrial signals simultaneously are now well understood.
There is a present proposal to deliver more localised services from the terrestrial repeaters. The National Association of Broadcasters in the US opposes this.
16.30 Mobaho in Japan and S DMB in South Korea – S Band GSO Plus ATC
Mobaho in Japan and S DMB had a similar network configuration but implemented at 2.6 GHz. A single geostationary satellite covers Japan with a left-polarised beam and S Korea with a right-polarised beam. A 12.226-GHz transponder on the satellite provides a downlink to terrestrial transponders to provided services in subways and tunnels.
16.31 Terrestar S Band in the US – GSO with ATC
Terrestar11 were granted access rights to two by 10 MHz allocations at 2000 to 2020 and 2180 to 2020 MHz. The band is shared with ICO. The launch of the initial Terrestar 500 spot beam geostationary satellite was originally planned for November 2007, but has been rescheduled for September 2008.
Terrestar have a joint venture with Orbcomm12 who have a legacy 30-satellite LEO constellation of micro-160-watt satellites with an uplink at 137–138 MHz and downlink at 400–401 MHz. Orbcomm went into Chapter 12 with Iridium and Globalstar but re-emerged and refinanced and at the time of writing provides services to M2M markets including telematics and asset tracking.
16.32 ICO S Band GSO with ATC
The FCC granted ICO13 a license for the other two by 10 MHz allocations at S Band with a mobile uplink between 2000 and 2020 MHz and a downlink at 2180 to 2020 MHz. An initial satellite launch is planned for the end of 1997 optimised for US coverage.
The ground-based components receive the satellite signal in K band, down convert to 2 GHz then transmit in synchronisation with the satellite signal. Additional satellites would provide improved inbuilding penetration.
The suggestion is that this network could provide up to 50 TV channels to mobile handsets and is therefore a potential competitor to Media FLO though the network would have other functionality including two-way voice and data, interactive multimedia and disaster-relief capabilities.
16.33 ICO S Band MEO at S Band with ATC
ICO planned a MEO constellation of 12 satellites in two inclined orbits at 10 400 km. The launch of the first satellite failed, but the second satellite went into orbit in June 2001 and is operational and capable of providing services to Africa, Europe and Asia. The ITU/CEPT S Band allocation to ICO specified a 12.5-MHz mobile uplink between 1997.5 and 2010 MHz and a mobile downlink from the satellites between 2187.5 and 2200 MHz.
There has been regulatory discussion as the continuing validity of ICOs claim to this spectrum. ICO base their claim to continued access rights on the basis of their legacy investments and stated plans to repurpose present and future satellites to support a DVB SH ATC network as a joint venture with Alcatel Lucent. Alcatel Lucent are leading a consortium with Sagem, Philips and DiBcom known as Television Mobile Sans Limite14 (Mobile Television without Limits) to promote wider adoption of the DVB SH standard.
China have several parallel standards initiatives using satellite and terrestrial interactive infrastructure and other localised reinterpretations of the T DMB and DMB T standards.
16.34 Eutelsat and SES ASTRA GSO – ‘Free’ S Band Payloads
Eutelsat15 has twenty three geostationary satellites. SES Astra16 has 36 satellites in 25 orbital locations. Together, they deliver over three thousand TV stations and over 2000 radio stations to 200 million cable and satellite homes. The transition to HDTV represents a challenge and opportunity to these providers and implies a possible future need to work more closely with the terrestrial broadcasting community.
There are no stated plans by either entity to implement a hybrid ATC network, however, an S Band payload wase included on the Eutelsat W2A satellite launched in 2008/9 and the companies have a joint venture working on delivering additional S band capacity.
The technology, engineering and launch costs of these platforms are already very adequately amortised across a substantial and largely captive subscriber base. S Band capacity can therefore be added at minimal incremental cost. It could be presumed that Intelsat would have similar economies of engineering, sourcing and subscriber scale that could be applied in a similar way. Other present and proposed S Band only providers need to consider the impact this could have on their future operational margins.
16.35 Intelsat C Band Ku Band and Ka Band GSO
Intelsat17 provides bidirectional transponder services to corporate and national governmental markets predominantly at C Band (3.4 to 7.025 GHz), Ku Band (10.7 to 14.5 GHz and Ka band (17.3 to 30 GHz). Intelsat has 51 geostationary satellites and uses spot beams to provide global, hemi, zone or spot beam coverage predominantly to fixed users. There are no stated plans for a hybrid ATC network. In common with Eutelsat and SES Astra, Intelsat has the benefit of multiple satellite constellation economies of scale and a large corporate subscriber base over which to amortise future S band engineering investments.
16.36 Implications for Terrestrial Broadcasters
As can be seen from the above, a number of these hybrid satellite terrestrial networks represent new competition for traditional terrestrial broadcasters. The ability to deliver TV to mobile users either from terrestrial repeaters or a satellite and/or simultaneously from both provides a measure of additional flexibility not available to terrestrial-only broadcasters. The addition of an uplink to support interactivity provides additional differentiation.
Conversely, most of the hybrid satellite terrestrial operators require access to terrestrial sites and terrestrial subscribers, many of which are largely under the control or influence of the incumbent traditional broadcast community. This suggests that collaborative rather than competitive ventures would be likely to yield better shareholder returns for all parties.
16.37 Implications for Terrestrial Cellular Service Providers
A number of these hybrid satellite terrestrial networks represent new competition for traditional cellular service providers. The ability to deliver full duplex voice, video and broadband data services to mobile users either from terrestrial repeaters or a satellite and/or simultaneously from both provides a measure of additional flexibility not available to terrestrial cellular operators.
16.38 The Impact of Satellite Terrestrial ATC Hybrids on Cellular Spectral and Corporate Value
Mobile satellite spectrum was largely allocated in 1997 and was either gifted or acquired on advantageous terms particularly when compared with the sums subsequently spent by cellular operators on PCS, WCS and most recently AWS spectrum. This incongruity has allowed the satellite operators to refinance and revalue. The FCC ruling to allow ATC terrestrial repeaters has substantially helped in this revaluation process.
The regulatory intention is to encourage investment in satellite spectrum that has remained fallow for over ten years. Some US cellular operators might be minded to question whether this is in retrospect or prospect an equitable arrangement.
Table 16.13 shows the IMT S Band cellular and satellite allocations in Europe.
Table 16.13 IMT S Band allocations
Table 16.14 ITU frequency bands including V Band
16.39 L Band, S Band, C Band, K Band and V Band Hybrids
The available radio frequency spectrum carries on beyond Ka band into V band, the millimetric band between 30 and 300 GHz. These are shown in Table 16.14 with a summary of allocations at lower frequencies.
These V band networks are ambitious satellite-based bidirectional broadband communication projects and include an advanced global EHF satellite hybrid MEO GSO network18 providing broadband connectivity for US Stealth Bomber aircraft.
This network is being implemented between 36.1 and 51.4 GHz and is a dual constellation hybrid combining a MEO constellation for low latency exchanges with a constellation of geostationary satellites for less latency-sensitive uploading/downloading.
Participation in these major military projects allows US vendors in particular to amortise engineering investments that can be translated to civilian applications. The EHF project suggests that future commercial networks may combine terrestrial repeater coverage with hybrid MEO and GSO satellite coverage.
16.40 Summary
There have been rumblings of discontent in the cellular and terrestrial broadcast community that new generation hybrid satellite terrestrial networks are really terrestrial networks with an ancillary satellite component rather than satellite networks with an ancillary terrestrial component.
These networks are being deployed into spectrum that was acquired at a fraction of the cost of recent (past 7 year) cellular spectral investments. A number of the operators have also emerged from Chapter 12 administration with engineering and launch investments largely paid for by their creditors rather than consumers. This would seem to confer an unfair advantage to these companies.
This may of course be true but pragmatically there will be a high level of interdependency between satellite /terrestrial ATC networks, terrestrial broadcast and terrestrial cellular networks both at infrastructure and subscriber level. Interdependency implies collaborative profit opportunity. More tellingly, a number of the propositions will have to be closely coupled with already-established terrestrial service providers in order to achieve long-term financial viability.
In particular, the significant economies of scale available to the cellular industry effectively dictate the radio functionality included in any handset. Satellite operators need to consider this factor with particular care. The positioning of Eutelsat and SES Astra is potentially advantageous given their substantial existing satellite and subscriber assets and ability to amortise engineering and marketing investments over a large and secure existing revenue base. Intelsat has potentially similar advantages. S Band payloads on any of these satellites are essentially free.
This factor combined with other positive satellite technology and cost trends including lower launch costs, larger more efficient pay loads and improved uplink and downlink performance suggest that satellites will play an increasingly positive role in future terrestrial mobile broadband service provision.
At the very least, satellites need to be more actively factored into present and future cellular and terrestrial broadcast business planning. Telstar marked the start of a new technical and commercial era in telecoms.19 Sixty years on it feels that the story has only just begun.
1 http://www.tiaonline.org/standards/technology/project_25/.
3 http://www.etsi.org/website/Technologies/TETRA.aspx.
5 Iridium is the 77th element and/originally there were to be 77 satellites in the constellation. Most Iridium found on the earth's surface is from meteorites.
6 http://www.iridium.com/about/IridiumNEXT.aspx.
7 https://www.siriusxm.com/player/.
8 http://www.siriusxm.com/servlet/ContentServer?pagename=Sirius/CachedPage&c=Page&cid=1018209032790.
9 http://www.tvtower.com/xm-radio.html.
10 http://www.mwrf.com/Articles/Index.cfm?Ad=1&ArticleID=5892.
14 http://www.generation-nt.com/alliance-alcatel-archos-pour-la-tv-mobile-dvb-h-en-bande-s-actualite-40220.html.
15 http://www.eutelsat.com/home/index.html.
16 http://www.ses-astra.com/business/.
18 http://www.satnews.com/stories2007/4128.htm.
19 Incidentally, July 2011 was the 50th anniversary of Yuri Gagarin's flight almost coinciding with the last planned mission of the American Shuttle.