The BBC Blogs - Spaceman
« Previous | Main | Next »

A Herculean effort to deliver broadband by satellite

Jonathan Amos | 12:24 UK time, Saturday, 18 September 2010

The date was September 1999 and banker David Williams was sitting on a beach in Santa Monica:

"I'd just spent a soul-destroying day at a satellite manufacturer, trying to push forward a project and getting bogged down in just the most ridiculous bureaucracy. And I was thinking there had to be an easier way of doing the satellite business. It's not that complicated - you get some money, you pay someone to build a satellite, you launch it, you flog the capacity. How hard can that be? I was venting my frustration to my wife and she said: 'if you think you're so bloody clever, go and do it yourself!'"

Roll forward to September 2010 and David Williams, now CEO of Avanti Communications, is just about to see his first satellite go into orbit on an Ariane 5.

Engineer works on Hylas telecommunications payload

The payload for Hylas was developed through Esa's Artes telecoms research programme

Hylas-1 is intended to be the first of three spacecraft (and more). Its primary market will be those people who live in internet-deprived areas - locations where terrestrial technologies such as fibre do not reach.

In the UK alone, the number of households that currently struggle to get a 2Mbps connection runs to many hundreds of thousands. Across Europe, it runs to many millions.

Hylas operates predominantly in the Ka-band, a part of the satellite-apportioned radio spectrum that allows high data rates.

Consumers on Hylas should enjoy speeds up to 10Mbps. In other words, it will help meet the Universal Service Commitment in the UK to provide 2Mbps to everyone by 2012.

The endeavour has been a private-public one. The British government put funding into the European Space Agency's telecommunications research programme, Artes, to produce the Hylas payload at EADS Astrium.

This innovative piece of kit can vary the amount of power and bandwidth needed to match the peaks and troughs in demand for net access across its European "footprint". David Williams:

"This is a big event for Britain. This is the first Ka-band system launched in Europe. It was done through a combination of Avanti, Astrium, and the British government just buccaneering in a really creative way to get something done. It didn't involve a huge, expensive, complex government programme - it was very efficiently done by the UK Space Agency, as it's called now, just deciding that Astrium should develop this technology and trusting Avanti to go and raise the money in the market to do it."

Today, Avanti has market capital of more than half a billion pounds. Its second satellite is fully funded and should launch 2015.

An artist's impression of Hylas in orbit

Hylas-1 is due to launch on an Ariane 5 rocket in November

The whole proposition is popular with investors. Just look at how its share price has performed... and remember, Avanti has yet to get its first satellite in orbit.

That doesn't happen until late November.

There are naysayers, of course. One of the most frequently heard concerns is that the broadband-by-satellite market will eventually be beaten back by terrestrial technologies.

Henrik Nyblom covers the telecoms beat at investment bankers Nomura. He described to me the position held by established satellite operator SES and some others:

"There are different views from within the industry. SES say they don't believe in a dedicated satellite; they don't think it makes sense based on the fact that terrestrial broadband will build out into those rural areas that today you can see satellite broadband being used. There is only a window of a few years where this service could really have a market and then terrestrial broadband - wireless or fixed - will fill these holes. SES argues that if you look at the auctions coming up for analogue TV spectrum that will be repurposed, such as in Germany, [the authorities] put as part of the criteria that to bid for this spectrum and to win this spectrum, you need to start building out rural areas before you start building out urban areas."

Williams' response is that some of the new terrestrial technologies such as 4G will operate data limits that no home user will find satisfactory:

"You cannot do fixed broadband substitution on a mobile phone network. A mobile phone network cannot supply the bandwidth that you need as a consumer to do broadband at home. What mobile phones do is mobility and they do that very well. The mobile phone companies will stick to mobility. Stephen Carter said that in the Digital Britain report last year. He said mobile phone technology is not the answer for fixed broadband, so anyone who says it is has either failed to consider, or understand, the technology, or has an agenda."

One established satellite player certainly shares Williams' position, and that is Eutelsat. The Paris-based operator has its own broadband-dedicated spacecraft launching not long after Hylas.

Called Ka-Sat, the spacecraft is something of a gorilla in space terms compared to Hylas.

Whereas the UK platform weighs 2.5 tonnes, has eight spot beams and can serve up to 300,000 customers; the French bird weighs almost six tonnes, has 80 spot beams and can serve more than a million customers.

One might fear for the nimble Brit with such a heavyweight coming on to the scene, but there is another widely held view in the industry that the pair need each other to drive open the market. David Williams again:

"I share that view. It was good news for us the day that Eutelsat announced that they were building a Ka-band satellite because it just added credibility to what we were already doing; and two operators competing in the market place will create more noise, more PR. The consumer will have some choices. The consumer will be able to compare and contrast, and that makes for a healthier market. And it probably means the market will expand faster."

Journalists like a conflict narrative, and it makes for some fun that the British and the French should be going into space to do the same thing at broadly the same time.

One thing seems clear - the consumer is going to be a winner; and making a minimum 2Mbps broadband service available to all European households has got to be good for the continent as a whole.


or register to comment.

  • 1. At 02:27am on 19 Sep 2010, Tom wrote:

    As I sit here, in Hong Kong, with a 1,000Mb/s (synchronous) fibre connection that is currently costing me about ten pounds per month, I can't help but wonder how much my compatriots back in the UK will end up paying for their satellite-delivered 10Mb/s.

    And as for why Mr Williams thinks mobile network operators will cap usage of data over their networks at an unpalletably low threshold, I suppose this must be a quirk of British business. For those parts of Hong Kong as yet too remote to enjoy fibre, the mobile operators, here, offer devices (USB 'dongles' and portable wi-fi hotspots) that support up to 21Mb/s with uncapped monthly usage for around 50 pounds per month, or 7.2Mb/s for around 20 pounds.

    Complain about this comment

  • 2. At 08:05am on 19 Sep 2010, powermeerkat wrote:

    10-20 Mb/sec connections (hardly a broadband) have no future.

    As HDTV, TV-based HD movie rental companies, etc. and wireless home concepts come on board, many a European country invests in a 1 Gb fiber networks.

    And not only in 'Old Europe' but in 'New Europe' as well.

    Look for example at Poland.

    As for 2Mb/sec - that surely must be a XX century joke.

    Complain about this comment

  • 3. At 08:59am on 19 Sep 2010, drj93gs wrote:

    It would be good to have this over the Oceans. Lots of market there - especially oil and gas.

    Complain about this comment

  • 4. At 6:54pm on 19 Sep 2010, unconvinced wrote:

    This is an interesting story. I don't know if Mr Williams has good long term market, but where I live in the country broadband coverage is only possible via expensive annual contracts. The dongles using mobile signals are utterly useless. They just aren't strong enough to provide a reliable connection, or any connection at all, usually. The old dial up, pay as you use, system is not offered for broadband (not enough profit for the suppliers, I suppose).

    I would like the option of reasonably priced broadband access on a pay as you use system, without the need for committing to a contract. If Mr. Williams could provide that, I'd certainly buy his product. As long as he is using private money to set up, I hope he succeeds. If he does everyone gains. If he doesn't the taxpayer doesn't lose.

    Complain about this comment

  • 5. At 8:03pm on 19 Sep 2010, jr4412 wrote:

    Jonathan Amos.

    "In the UK alone, the number of households that currently struggle to get a 2Mb/s connection runs to many hundreds of thousands. Across Europe, it runs to many millions."

    and that is why we ought to press for fibre-optics being laid everywhere, they're cheap to produce (and support much higher bandwidths both ways) but require loads of manpower to install; right now most of our economies are buckling under a growing number of unemployed and laying cable would employ many thousands. satellites are more 'cool' but in real terms, their manufacture and deployment benefit only a relatively small number of 'stakeholders'. as you say, that "..proposition is popular with investors.." but where's the benefit to wider society?

    Complain about this comment

  • 6. At 9:12pm on 19 Sep 2010, Jonathan Amos wrote:

    There's a section in the Digital Britain report which is worth recalling. It draws the distinction between two separate but related issues - one is making sure the current generation of broadband gets to everyone, and the other is about the ability of the UK to roll out the next generation of broadband that is able to deliver speeds of 100Mb/s and above. The report said: "These are two quite distinct national undertakings. Universal availability of today’s network essentially requires incremental upgrades of existing infrastructure and the costs are therefore limited to the hundreds of millions of pounds. Delivering tomorrow’s network essentially involves installing a new network or networks, and the costs are in the billions." Much of what I talk about above relates to the universal availability of today’s network. Whilst we can all have an interesting debate about the need for fibre, how long it should take to install and who should pay - it shouldn't be forgotten that there are still a great many people out there for whom 2Mb/s is still a dream.

    Complain about this comment

  • 7. At 10:09pm on 19 Sep 2010, TeamYankee wrote:

    As an IT professional who has had to support individual users using sat broadband it's a bit of a joke for any "interactive" usage.

    The end to end delay is on the order of 500ms to 900ms or more. For surfing the web it's bearable, but for anything else it's unuseable.

    The transmission delay to geosynchronous orbit and back is major factor.

    You can forget online gaming or using it for occasional business use it's way too slow.(Citrix, terminal server etc.)

    Complain about this comment

  • 8. At 11:28pm on 19 Sep 2010, 8138 wrote:

    We in the rural areas would love to have broadband at any speed.
    There are lots in our community who can still only use dial up.
    The ones that do have broadband are lucky to get 0.5 meg and that can be off for hours at a time.
    All this talk of high download speeds makes us seem nearly prehistoric.

    Complain about this comment

  • 9. At 11:45pm on 19 Sep 2010, jr4412 wrote:

    TeamYankee #7.

    "..using sat broadband it's a bit of a joke for any "interactive" usage."

    yes, but 'received wisdom' (dogma?) will see our money wasted by prioritising "..the universal availability of today’s network" (Jonathan Amos #6), ie adding satellites.

    as the old saying goes, you can take a horse to the water but you cannot make it drink.

    Complain about this comment

  • 10. At 5:45pm on 20 Sep 2010, Paul_Essex wrote:

    Just a few points.
    #1 Tom - mobile phone operators are already stopping unlimited data packages (O2 announced this in June 2010). It all comes down to physics - there is only so much capacity to go round! Not only that, there are places here in London where you struggle to get a signal, let alone on line!

    #5 JR4412 - You've just summed up in a nutshell why satellite broadband has a future! Compared to the cost and disruption of laying fibre to tens of thousands of rural houses in the UK, building a satellite is cheap. This isn't aimed at replacing broadband for people withing a few miles of a switch, its for the many many thousands of people who are really rural. I live within 30 miles of London and the best I can get via a phone line is 4mbs - what hope would someone living in the middle of Lincolnshire, Scotland or Wales? Also, look at the opportunity in less developed countries. People gain such an advantage through having a decent internet connection (shopping, entertainment, education, commerce, communictaion) that I applaud anything which will help bridge the digital divide!

    In my view there is not a 'one size fits all' solution to getting everyone on line. It's about using a mix of available technologies (i.e fibre, mobile and satellite) and satellite can provide to those areas where the others cant (or wont because of it's not financially worth their while).

    Complain about this comment

  • 11. At 7:10pm on 20 Sep 2010, skeptic456 wrote:

    The problem is, as TeamYankee pointed out, latency. It's not the speed of the connection, it's the amount of time it takes to get a connection and the time it takes to send anything down that connection.

    The satellite will be a geostationary orbit which means it'll be about 35,000km out in space. To get a response will require two round trips, there and back to get to the target server and there and back again to send the response. So from initiating a request to receiving the response involves data traveling about 140,000km. The speed of light is about 300,000km/s so, even forgetting any other latency or processing, the best case is about 500ms between doing something on your computer and seeing the response.

    This will be fine for streaming media (TV, audio, etc.) or downloading large files because it just involves the remote server pumping data down the connection once it's established. But anything that involves making lots of small requests (e.g. online games, browsing the web or the sorts of things you'd normally do over a corporate VPN) will be virtually unusable.

    Complain about this comment

  • 12. At 9:52pm on 20 Sep 2010, otura78 wrote:

    I currently live in England where I get within the M25 around 12 miles from central London and also for part of the year in Spain near Granada where I get 5mbps. However the cost is around €10 in England and €40 in Spain. I get the slow speed in England because we live around 4 miles from an exchange and there is no cable and BT will not put boosters in. In Spain it is great but the cost is terrible and we cannot get mobile broadband because there is no signal in our valley. I satellite might do it then for both locations....

    Complain about this comment

  • 14. At 12:14pm on 21 Sep 2010, unconvinced wrote:

    Excuse me for posting twice but this issue fascinates me. If the satellite venture takes place would it provide for a one global isp arrangement? I have a big problem in that I travel to various locations, in different countries, where dial up is too slow but dongles are useless. If I have a contract with a fast connection company in one country, I can't use it in the other countries. This costs me a fortune ( or would do if I paid it).

    If I had the option of a universal, reasonably priced, isp which I could access irrespective of national location, I would love it. I think some of the doubters who have posted here assume that all broadband users demand extremely high speeds. Is that true? Not everyone wants to play games or needs massively high speed connections.I believe there might be a market for the kind of system provided by Mr. Williams's sytem, which for me would give perfectly satisafctory speed. I'd certainly be interested.

    Complain about this comment

  • 15. At 4:27pm on 26 Sep 2010, wayne wrote:

    Broadband is great. But is satellite service broadband? It's an always on connection, but the speed is poor and Hughes and WildBlue services both have fair access policies (FAP) that cut your speed to 64k when you exceed the download upload. Hughes cuts VPN speeds to 64k period. I suspect WildBlue and all the others are the same... so you can not generally work from home. pity.

    Complain about this comment

View these comments in RSS


Sign in

BBC navigation

BBC © 2014 The BBC is not responsible for the content of external sites. Read more.

This page is best viewed in an up-to-date web browser with style sheets (CSS) enabled. While you will be able to view the content of this page in your current browser, you will not be able to get the full visual experience. Please consider upgrading your browser software or enabling style sheets (CSS) if you are able to do so.