Latest News

Ka-band: Cautious Optimism

By Staff Writer | June 1, 2003

By Peter J. Brown

Throughout the next 18 months, commercial Ka-band satellite capacity will grow considerably. New North American broadband multimedia ventures are starting to talk about specific plans rather than a wide range of opportunities. In Europe, however, the initial phase of this trend is almost over.

Military use aside, Ka-band must be seen as more than just a frequency, as spotbeams, phased array antennas and new software technologies entering the picture. Because the hype is gone and the mood is more cautious, yet optimistic nonetheless, there is no real audience for any wild-eyed projections. Most if not all of the early backers with their deep pockets also are gone.

With more than 80 Ka-band geostationary orbital slots assigned to U.S. operators by the FCC, and with almost 50 active licenses assigned in two rounds, there is simply not enough space here to update the status of all the Ka-band ventures, both geostationary and non-geostationary systems.

Still, look at all the recent and quite unexpected developments. Echostar Communications Corp.’s decision to acquire Ka-band capacity on SES Americom’s AMC 15, and Liberty Media Corp., Intelsat Inc. and the National Rural Telecommunications Cooperative (NRTC)’s decision to invest in CO-based Wildblue Communications are signs of forward momentum adapted to the new realities of the broadband arena.

Besides obtaining a stake in Wildblue, Liberty Media reached an agreement with Lockheed Martin Corp., Northrop Grumman Space and Mission Systems Corp. (formerly TRW Inc.) and Telespazio S.p.A. involving a proposed restructuring of Astrolink International LLC. Liberty Satellite will acquire substantially all of the assets of Astrolink, which is scheduled to initiate service in late 2005, while the two satellites for Astrolink are simultaneously back on track at Lockheed Martin and Northrop Grumman.

"We here at Astrolink are not familiar with the specifics of Wildblue’s market strategy," says Steve Soroka, Astrolink’s vice president of operations. "That said, given our performance levels, we believe that our service offerings are complementary with Wildblue’s. Our services are directed at the enterprise market, and include bandwidth on demand and Virtual Private Networks (VPNs), as examples."

What does the NRTC, for example, see as the value proposition in this transaction with Wildblue? "Our members will be poised once again to deliver a valuable product that is needed and desired by their customers in rural America," says Roy Heffernan, NRTC’s senior vice president of business development. "The ability to deliver a converged data and video product, even potentially bundled with their electric or telephone services, will put our members in a strong strategic position.

"NRTC is thrilled with our new partners, Liberty Satellite and Intelsat, as well as the existing team of investors, and we look forward to maximizing the return on our collective $156 million investment," adds Heffernan.

In Europe, which emerged as the true launch pad for the global Ka-band sector, there are other indicators of changes underway. Eutelsat is now exploring the merging of its Skyplex platform with the Ka-band capabilities on its new Hot Bird 6 satellite to enable micro-broadcasting as one of many possible interactive products, while SES Astra plans to use Ka-band capacity on its Astra 1H for a new always-on satellite return path-based service known as Satmode, short for satellite modem. SES Astra links up with partners Canal Satellite, Canal+ Technologies, Newtec, STMicroelectronics and Thomson in this venture. Satmode is seen as a thin-stream, interactive solution that will boost demand across the board for various new gaming, video-on-demand and messaging services.

As for Ka-band in Asia, Space Systems/Loral completed static load testing of iPSTAR 1, the 14,900 pound 1300 satellite under construction for Shin Satellite Plc in mid-April. With 84 spotbeams, this hybrid Ku-band/Ka-band platform will be operational at 120 degrees E starting in 2004. iPSTAR 1 will join other Ka-band satellites in the region, including New Skies Satellite NV’s NSS 6 at 95 degrees E, Koreasat 3 with three Ka-band transponders, and the SCC Superbird and JSAT NStar satellites serving broadband markets in Japan.

"The market is certainly much smarter now and companies cannot justify the deployment of large Ka-band networks. We will not see much demand for these services before early 2005," says Chris Baugh, president of Northern Sky Research in Orland, FL. "Incremental business, with reduced upfront costs, is where the emphasis lies."

According to a recent Northern Sky study of Digital Video Broadcasting-Return Channel Satellite (DVB-RCS) worldwide, there are an estimated 32 hubs and 8,500 terminals in operation today. Baugh reminds readers that few, if any, of these hubs and terminals are using Ka-band capacity, with the exception of a handful that may have a Ka-band return via the SES Astra 1H satellite. Otherwise, the vast majority of DVB-RCS deployments to date are Ku-band only.

"You have a delicate situation here. None of the operators can afford to be late because they risk the fact that their brand will not be associated with this new technology," says Karim Nour, an analyst at Frost and Sullivan. "At the same time, if they come too early to either the enterprise or consumer markets, they will find nobody to sell Ka-band to."

Although subscriber acquisition costs (SAC) will not represent a big hurdle in the enterprise sector, consumer business models are a different matter altogether, according to Nour, because they will need to accommodate a SAC much higher than the one associated with mainstream DTH services, due to significantly higher terminal costs.

"How do you stimulate a broader consumer market? Not through government subsidies. You have to hit a certain price point, a very specific dollar amount in the $40 to $50 per month range," Nour says.

Sorting Out The Standards

While much has been said about DVB-RCS and Data Over Cable Service Interface Specification (DOCSIS) over satellite, in terms of their role as the two fundamental standards that will serve as the foundation for next generation satellite-based broadband multimedia services, there are other standards in motion as well.

The Spaceway Air Interface from Hughes Network Systems (HNS) is making steady headway too, and there are likely to be other options in the future.

"We also have a placeholder for additional families, such as a EuroSkyWay family, but the current state of the economy means that this work is currently on hold," says Rupert Goodings, president of U.K.-based Ecotel Ltd., and an independent consultant representing HNS for ongoing work on standards at the ETSI TC-SES (Technical Committee-Satellite Earth Stations) and Systems Working Group on Broadband Satellite Multimedia (WG-BSM), which he chairs. Goodings is also vice chair of TC-SES and chairman of another SES working group addressing mobile services, GMR-WG.

"The BSM concept is to divide the air interface into two parts, consisting of the upper satellite independent part and the lower satellite dependent part, which is partly based on the concept from the European IST-BRAHMS project," Goodings says.

The Spaceway technology is being standardized as a BSM family (i.e. lower layer set of standards) under the name "RSM-A," which stands for Regenerative (satellite payload) Satellite (return channel) Mesh topology. A DVB-RCS family with the name "TSS-A", or Transparent (satellite payload) Satellite (return channel), Star topology is evolving as well.

By focusing a major new effort on developing common upper layer standards that directly address the issues of interworking IP over satellite, the WG-BSM is confronting several overlapping issues head-on, such as the fact that in terms of multicasting, IP standards do not even address multicast and broadcast. Transmission Control Protocol (TCP) enhancements are part of the mix too, with the intention of improving performance through long delay paths.

"The main challenges are at the higher layers, which have to deal with the satellite unfriendly aspects of the IP protocols. The WG-BSM is attempting to focus on this aspect, i.e., common solutions to reduce IP signaling overheads, increase TCP performance, etc.," Goodings says. "These solutions should be independent of the satellite dependent lower layers (modulation, coding, etc.), and should be solutions to shared problems such as delay and bandwidth limitations.

"When it comes to Quality of Service (QoS), for example, IP QoS standards are not designed with satellite links in mind," adds Goodings. "I think proprietary software solutions to these problems are what lead to any perception of user unfriendliness. Such solutions must constantly be upgraded as the IP standards themselves evolve. What we really need are standardized approaches to the satellite independent layers, which take into account forwards and backwards compatibility, as well as more uniform interaction with the IETF (Internet Engineering Task Force)."

An Unexpected Twist

The European Ka-band rollout is unfolding slowly and essentially on schedule, despite the unfortunate loss of Astra 1K after a launch misfire a few months ago. Luxembourg-based Satlynx S.A., an SES-Global and Gilat company, is adjusting its game plan accordingly, according to Robert Feierbach, vice president of corporate and DTH services at Satlynx.

"The loss of 1K has forced us to fine tune the existing Ka-band beam configuration on 1H. This involves a slight change in the slanting of the eight beams in order to fill up the gaps in our key markets in France, Spain, Italy, Germany and the United Kingdom that were to be addressed by 1K," says Feierbach.

The Ka-band DVB-RCS Satellite Interactive Terminal (SIT) market is moving ahead with Montreal-based EMS Technologies Inc., as well as Newtec, Nera and Raytheon. Current prices are in the 2,000-3,000 Euro (U.S. $2,160-3,240)-plus range. Satlynx projects a non-cable, non-ADSL target market in Europe of approximately 250,000 corporate sites, and 1.3 million Small Office/Home Office (SOHO), "prosumers" or small-to-medium enterprises. The market will expand with some 7.3 million consumers, who will only become accessible in 2005 once SIT prices fall to under 300 or 400 Euros (U.S. $324-432).

"Here in Europe, most service providers are leaning toward DVB-RCS, although DOCSIS is under consideration as well. For our Broadband Interactive (BBI) service, we use DVB-RCS for the high-end corporate market. With our proprietary Gilat interface, starting with the 360E family, up to four or five PCs per terminal can access our services," says Feierbach.

In addition to corporate LAN clients who are billed on a per gigabyte basis, a hybrid satellite/Wi-Fi consumer-oriented platform is emerging as a top seller, particularly in Spain and Italy. "Many people in North America probably do not realize that Spain has taken such a lead role. Wi-Fi is very hot in Europe, and these rural systems costing under 1,000 Euros (U.S. $1,080) can cover an area with a radius up to 1 km," says Feierbach. "I estimate that 40 percent of our Spanish BBI installations to date involve Wi-Fi networks."

Other than the high-end "prosumer," or SOHO market, Satlynx has no consumer market at this time. Satlynx is adding more network hooks, customer visibility, accessibility and tools for customers accessing their hubs. For example, reference sites for each beam have been put in place to enable customers to troubleshoot QoS across the eight beams. The goal is to simplify the terminals by not adding bells and whistles, while commercial hub features continue to evolve.

North America Gets Ready

The launch of Nimiq 2 by Telesat Canada signals the beginning of a new era in North American satellite ventures, as a significant amount of commercial Ka-band capacity lights up. With Echostar 9’s launch date in mid-year and, later in 2003, Anik F2 preparing to head skywards, the stage is set for a new range of broadband multimedia services.

Anik F2, in addition to C- and Ku-band capacity, will carry 45 Ka-band circular spotbeams, with 30 licensed to Wildblue Communications and 15 dedicated to Telesat’s Canadian customers. Total throughput will range between 3 and 4 Gbs.

"Our success will be determined by how well we price and package our terminals. We have an ultimate target of less than $500 for the consumer market, and we have a Ku-band terminal for under $1,000 in mind as our target in the enterprise market, which is our primary objective," says Paul Bush, Telesat Canada’s vice president of business development. "Telesat has RFPs (Request For Proposals) out to vendors for the building of three gateways and terminals in Canada. Our mission here is to build volume and drop prices."

Brad Greenwald, spokesman for Wildblue, said his team was keeping things close to its vest, and the recent decisions to invest by Liberty Media, Intelsat and the NRTC, among others, is opening up a new chapter as the company prepares to win over an estimated 30-35 million unserved residential and SOHO customers.

"A number of milestones, including the granting by the FCC of a change of control motion in March, have been met and we are now working on the paperwork to officially close the deal," says Greenwald.

There is no firm launch date for Wildblue 1, which is to follow Anik F2. Work on gateways by Andrew was restarted after being put on hold several months ago and both DOCSIS and DVB-RCS interface standards were on the list.

"Our belief is that in order to attract a consumer market, you need to charge less than $50 per month, and that CPE [customer premises equipment] prices have to come down to a couple of hundred dollars," says Greenwald.

On Anik F2, a Boeing 702, a half dozen wideband 492-MHz transponders will beam traffic to the six planned gateways–three each in the United States and Canada–relaying multiple MF-TDMA return carriers from users grouped into six to eight beams in the process.

The F2 gateways use a series of 56-MHz transponders powered by thirty-one, 90-watt TWTAs to access users in each spotbeam, with 17 spotbeams designated as heavy rain beams for customers in the eastern half of North America with a single 90-watt TWTA assigned to each beam. The other 14 TWTAs support a pair of beams each.

BeamLink, described as an advanced, flexible multiplexing subsystem, can be adapted to meet the requirements of numerous different traffic patterns amongst users in each group of beams, while the SpaceMux onboard processor will demodulate uplink MF-TDMA carriers, remultiplex the packets, and then remodulate the TDM downlink. Bear in mind that achieving true five to six times frequency reuse demands adequate spatial isolation at all times. The spotbeam coverage must also cope with the ever present possibility of rain fade, which will require variable uplink power at both the terminal and the hub.

"We want to try and do most of the adjusting at the hub and we will use advanced techniques to ensure continuous operations, even with rain fade, by stepping down speeds of operation if conditions warrant it. That means taking the return channel speed at 500 kbs and easing it back to 384 kbs or even 194 kbs," says Bush. "The power levels on F2 are 55-58 dBW. These power levels, coupled with adaptive coding, will enable us to provide a competitive Internet service to consumer and enterprise customers."

Game Plans Emerging

Both SES Americom and Loral Skynet which has now absorbed its Cyberstar unit, are pursuing the Ka-band market, but both seem to be headed in very different directions in the process. SES Americom has successfully leased its entire block of Ku-band capacity, along with one polarity of the full Ka-band payload on AMC 15 to Echostar Communications Corp.

Prior to December 2003, Echostar plans to use its new Space Systems/Loral-built Echostar 9 to start testing the Ka-band waters. Echostar would not divulge exactly what sort of testing would go on. At the same time, the status of its Ka-band Visionstar project is uncertain, as Echostar’s request for an FCC extension has yet to be approved.

AMC 15, with its payload of two dozen 36 MHz Ku-band transponders and a dozen 125 MHz Ka-band spotbeams, is scheduled for launch in August 2004.

"This validates our concept of Americom2Home (A2H). We are essentially real estate developers and we have just got a very large tenant in our first building," says Kevin Smythe, senior vice president of residential satellite services at SES Americom. "We are in the process of modifying both AMC 15 and AMC 16 to make them more DTH-friendly. We have characterized A2H as replicating what SES Astra has done already in Europe with respect to wholesaling capacity to DTH providers."

While there is no firm launch date for Telstar 8, a Ku-band forward channel/Ka-band return channel-equipped satellite, Loral Skynet expects it to be built by the end of this year. It will open the door to spotbeam-based services using DVB-RCS, or whatever standard a service provider decides to implement, according to Robert Hedinger, executive vice president of sales, marketing and client services at Loral Skynet.

"Our plan is to extend our Ka-band capabilities on Telstar 8 with additional Ka-band capacity as the market develops. We want to time our infrastructure rollout so that [it] is there when it is truly needed," says Hedinger. "We can start a wide range of IP-based and VPN services at Ku-band and migrate them to Ka-band when the time is right.

"We are frequency independent. We do not look at the market as strictly a Ka-band phenomenon. Instead, we are following the market in an incremental fashion. We see an evolutionary process unfolding in terms of implementing next generation VSAT solutions with broader dynamic bandwidth allocations and hubbed networking capabilities," Hedinger adds.

Strategies Abound

The satellite industry is pursuing hybrid networking strategies along with the rollout of Ka-band broadband multimedia services. We see this underway, for example, with Panamsat Corp.’s launch of its new virtual teleport concept involving the consolidation of six Panamsat teleports, and decision to flow Panamsat video and data traffic through Level 3 Communications Inc.’s Multi-Protocol Label Switching (MPLS)-enabled network. With this in place, Panamsat customers will now be able to access this hybrid network via Level 3 points of presence in 70-plus cities in the United States and Europe.

Path 1 Network Technologies Inc., based in San Diego, CA, is providing its Cx1000 IP video gateway technology for this hybrid network. "Panamsat is using the Path1 CX1000 IP video gateways as part of the virtual teleport project to interconnect our teleports for transport of broadcast-quality video over fiber networks," says Bridget Neville, Panamsat’s senior vice president of engineering. "We are also using a Tandberg 6120 to provide similar functionality to the Path1 product. The video over IP technology is geared toward contribution of broadcast-quality video. The video over IP technology is relevant to our ground infrastructure (fiber and teleports), and is therefore agnostic to the satellite frequency, whether C-, Ku-, Ka-, or even X- band."

Path 1’s solution was demonstrated recently by ND Satcom, which used the video gateway to encapsulate and encode a high quality video signal via an ASI port into the fast Ethernet port of the ND Satcom SkyWAN system.

"Such a device is always required if a video signal is available via an ASI port and needs to be transferred via the Broadband Media Network solution. Our solutions work well with Ku-band, C-band and Ka-band geostationary satellites," says Peter Neu, manager of product marketing at ND Satcom. "The questions surrounding Ka-band involve cost, which is expected to come down, especially for the ODU (Outdoor Unit), and availability which is expected to be worse than with Ku-band because of rain fade. So, it will depend on customer requirements, applications and the specific rain zone where the customer is located.

"Today, most Ka-band projects are dedicated to the residential and SOHO business. Our approach is targeting high performance, professional applications for broadcasting and media companies. High quality and reliability is key, so Ku-band is adequate," adds Neu.

The emerging broadband multimedia marketplace will also be impacted significantly by advances in MPEG-4 encoding, and solutions that transcode pre-compressed MPEG-2 satellite video channels to H.264-enhanced MPEG-4 aka MPEG-4, Part 10.

For example, Santa Clara, CA -based iVast paired up with VideoTele.com, a division of Tut Systems Inc., to offer an MPEG-4-based digital headend. iVast has devoted considerable man hours to developing standards-based, interactive digital media solutions.

"Yes, iVAST’s MPEG-4 solution will match MPEG-2 video quality at a lower bandwidth. This can yield two times or three times bandwidth savings, or broadcasters can use the same bandwidth to deliver higher quality video such as HDTV," says Alex Derecho, vice president of professional services at iVAST.

By taking advantage of MPEG-4’s inherent compression efficiency and system layer support, iVAST is focused on enabling the delivery of high quality, rich media with previously unobtainable bandwidth efficiency, according to Derecho. "We expect broadcasters to migrate to MPEG-4. Our indications are that they’ll likely do so to address the bandwidth issues associated with delivering HDTV first," Derecho adds. "The iVAST product suite will enable both increased channel density and the ability to offer more compelling services on existing Ku-band platforms."

These changes are coming quickly and furiously. Broadband multimedia via Ka-band and new terrestrial wireless breakthroughs, including Wi-Fi, video over IP and MPEG-4/H.264, unite under the banner of content delivery, whether the content in question arrives in real-time or not.

We also urge our readers to keep in mind that one of the strong selling points for Ka-band is the supposed shortage of Ku-band, among other things. To the extent that new digital encoding platforms and compression schemes, along with advances in video over IP free up Ku-band capacity, this will impact the demand side of the satellite transponder market as a whole, and it could reshape broadband multimedia strategies in a dramatic fashion going forward.

In a nutshell, how much goes up in the sky, and how much stays on the ground will certainly help to determine the future of the new cluster of Ka-band satellite broadband multimedia ventures.

Go online to read more about what is unfolding in the emerging Ka-band marketplace at http://www.viasatellite.com.

Peter J. Brown is Via Satellite’s Senior Multimedia & Homeland Security Editor. He lives on Mount Desert Island, ME.