Wireless Waffle - A whole spectrum of radio related rubbish
(Not) A Hope In Heavensignal strength
Tuesday 29 January, 2013, 12:34 - Spectrum Management, Satellites
Posted by Administrator
Following the recent Wireless Waffle post about Wes's excellent Soldery Song, it has been noted that the terms 'C-Band' and 'Ku-Band' were thrown around casually, as if everyone understands what these mean. For the benefit of those who don't, here's a short description of what they're all about:
  • C-Band (in satellite terms) refers to frequencies in the range of roughly 4 to 8 GHz. C-Band satellites are commonplace in areas where there is very heavy rainfall (eg tropical areas) as, using relatively low microwave frequencies, signals propagate better and suffer fewer rain fades. It's not the total amount of rain that matters, it's the voracity of the downpours.
  • Ku-Band refers to frequencies roughly in the range of 10 to 15 GHz (the 'roughly' refering to the fact that uplinks and downlinks work in different bands that may or may not be within this range but are at least somewhere nearby). Ku-Band is the band most commonly used for digital satellite television broadcasting anywhere that it doesn't rain too heavily (eg most non-tropical areas). In many areas, the geostationary arc is near saturation in Ku-Band, with satellites positioned every 2 degrees or so. Some satellite operators have tried to launch broadband internet services in this band but the high levels of congestion make it difficult to find the space to do so.
  • Ka-Band refers to frequencies roughly in the range of 20 to 30 GHz. Ka-Band is the relative newcomer on the block (though Italsat F1 conducted some tests in this band in the early 1990s). With Ku-Band becoming congested with broadcasting, satellite operators wishing to offer broadband services are increasingly turning to Ka-Band to do so and it is fast becoming the most hotly contested band for new satellite services. Already popular in the US, it is now being successfully commercialised in Europe and Africa. In emerging countries, where terrestrial (including wireless) broadband services are unlikely to be rolled-out any time soon, the use of Ka-Band satellite broadband is likely to open up Internet access to millions of people who wouldn't otherwise have a hope of getting a decent connection.

newsatlogoIt is perhaps no surprise that in the battle to gain access to Ka-Band, some of those doing the fighting are going to lose out. This seems to have happened to Australian satellite company NewSat. Recently both Morgan Stanley and Lazard have pulled out of the programme raising funds for the launch of NewSat's Jabiru 1 satellite. Digging into the economics of NewSat, it appears that the numbers just don't add up. The cost of their capacity would be around 50% higher than that of existing Ka-Band satellite operators, which might be a difficult sell when trying to raise lots of capital to build and launch a new satellite.

doesnt add upFrom the figures on their web-site, NewSat were going to pay Lockheed Martin US$550 million to build and launch their 'bird'. The resulting satellite would operate in a total of around 8.4 GHz of Ka-Band spectrum. Compare that to UK satellite company Avanti who paid US$400 million for their bird, but which accesses 11 GHz worth of Ka-Band spectrum. Do the maths and you see that NewSat would be paying approximately US$66 million to access each GHz of spectrum, whereas Avanti paid only US$36 million. This difference in cost is borne out in prices too, with NewSat claiming to be aiming to charge customers US$1.3 million for a transponder for 12 months. The equivalent price for Avanti is around US$0.9 million.

This may not yet be a Ka-tastrophe (groan) or even a Ka-lamity (double groan) for NewSat. NewSat suspended trading in its shares in November to give it time to re-think it's financing strategy. Maybe there is still a 'hope in heaven' for those communities that NewSat was intending to serve, if, of course, the other satellite operators don't get there first!
7 comments ( 2566 views )   |  0 trackbacks   |  permalink   |   ( 3.1 / 699 )

Arqiva win UK Local TV multiplex licencesignal strength
Monday 28 January, 2013, 12:22 - Licensed
Posted by Administrator
In a shock decision, Ofcom today announced that Arqiva are to run the UK's local television multiplex. The winning company, called 'Comux' has won the licence to operate the transmission network that will support local television services. Not speaking to Wireless Waffle, a spokesperson for Ofcom was overheard to think,
"We like Arqiva, oops, we mean Comux's proposal. After all, Arqiva transmit 'X-Factor' which brings us a lot of business in dealing with complaints: without Arqiva a lot of us would be out of a job. We also like the idea of supporting monopolistic companies that pay no corporation tax in the UK. We are hoping that Starbucks will bid in the 4G auction we are running at the moment and that Network Rail will bid to run a mobile network for trains (What? They already do? Since when?)"

jack hargreavesBut wait, in their public justification for the award, Ofcom state (page 6 of the account of decision),
Arqiva is not an applicant for the licence...

So is Comux really Arqiva, that is the question. Well they certainly share the same postal address: Arqiva/Comux, Chalfont Grove, Chalfont St. Peter, Gerrards Cross SL9 8TW.

OK, it's not exactly conclusive but the fact that they share an office surely means there are opportunities for them to share a lot more too! Collusion is a very dirty word, almost as dirty as tax avoidance.

Either way, here's raising a glass at local television in the UK. Let's just hope they bring back Old Country with Jack Hargreaves. What? You mean he's already dead? Since when?
1 comment ( 2909 views )   |  0 trackbacks   |  permalink   |   ( 3 / 918 )

Hop to it!signal strength
Friday 25 January, 2013, 13:45 - Radio Randomness, Spectrum Management
Posted by Administrator
Wireless Waffle received an e-mail from Des of Ireland. Des writes:
Since early May I have been noticing many many frequencies being occupied by very short bursts of digital 'noise' which are random in their frequency and time but very recognisable. So far pattern emerged is that they follow an 8 kHz spacing right across the HF bands (from 3.4 MHz to 28.5 MHz), but mainly in 6 to 9 MHz region. Even 6622kHz Shanwick being clobbered ... These noise bursts in the HF bands intrigue me, I wondered if it is a basic military comms set-up in case satellites/internet/microwave/fible-cable are clobbered.

Take a look a the picture below (click on it to open a much larger version). It is a snapshot of the radio spectrum between roughly 6550 and 6950 kHz taken using the University of Twente's on-line receiver in the Netherlands (which is a marvel in itself). The snapshot was taken at about 07:00 GMT. The horizontal axis shows the frequency, the vertical axis is time (in thie case about a minute). Straight vertical lines represent constant transmissions. Dotted ones (such as the broken line just above 6600 kHz) are morse code. Other squiggles that are roughly vertical are all manner of other signals that can be found on the HF bands.

hf frequency hopper

What is of interest here are the horizontal dashes of which there are three at the top left hand corner (just under 6550 kHz), four just below 6950 kHz and various others scattered across the chart, seemingly randomly (see around 6665 kHz and 6555 kHz for two bright ones). These are not bugs in the University's software, nor are they local interference in Twente. What they are are bursts of data from a frequency hopping transmitter. If you tune into one of the frequencies just at the time when the transmission is taking place on that frequency, you will hear a 'chuff' noise which is the quick burst of data that is being sent. If you happen across a frequency that has multiple 'hops' on it, the effect is not totally unlike there being a steam train on the frequency (listen to this actual recording).

At HF, this hopping transmission is almost certainly military in nature. Frequency hopping at HF is not at all uncommon. Even back in the 1980s, Racal's TRA 931XH would happily hop around the HF bands. In the case of the '931XH it did this by changing frequency roughly every second. Transmissions were just SSB (with an initial data burst to synchronise the receiver and transmitter - this is essential so that the two follow the same sequence of frequencies). The Wireless Waffle team had the fun of seeing a demo of the '931XH which was set to hop from frequencies between around 6950 and 7450 kHz, right across the 41m broadcast band. The effect of the hopping was to change the background noise every second or so - sometimes with a loud whistle caused by the carriers of the broadcast signals. The effect to anyone who happened to listen on a frequency that was being used would have been that they would have heard speech for a second which would then disappear.

hop to itThere's nothing unusual about the use of frequency hopping transmitters. Your bluetooth headset does this, and most GSM networks are set up to use frequency hopping too. The reason for using frequency hopping can be many and various, such as:
  • Hopping around makes the transmission much more difficult to detect. Unless you know the sequence of frequencies being used, it's almost impossible to follow the transmission from one frequency to the next.
  • Hopping can overcome some kinds of interference. If one frequency is blocked (from a broadcast transmission for example) the information sent on that frequency is lost, but if most are clear of interference, the error correction schemes can be arranged to deal with missing blocks and the overall communication is unaffected.
  • Hopping can help overcome fading and propagation problems. In a GSM network for example, Rayleigh fading will cause some channels to have deep fades and others not. Hopping around makes sure that these 'dead' channels do not cause a total lack of communication.
It's not surprising then that the military are using hopping on the HF bands (nor anywhere else for that matter). The question that remains unanswered is whether the military still need HF given all their other channels of communication. Patently they do!
1 comment ( 2536 views )   |  0 trackbacks   |  permalink   |   ( 3 / 732 )

What do coffee, books, fiddles and TV pictures have in common?signal strength
Wednesday 16 January, 2013, 03:58 - Broadcasting
Posted by Administrator
The UK public has been angered recently at the discovery that some high profile companies used various 'fiddles' to avoid paying tax. At a time when UK incomes have been squeezed, the idea that big name companies have not been contributing to the UK tax coffers left them hopping mad. The first company to be outed was Starbucks who, despite generating 400 million in revenues, paid less than three pence and a used muffin wrapper in corporation tax (corporation tax normally represents 20% of a company's profit).

Next up was Amazon whose sales in 2011 generated 3.35 billion yet who paid only 1.8 million in corporation tax. Even internet behemoth Google paid only 6 million in corporation tax against a turnover of 395 million.

These foreign companies largely argue that they pay tax elsewhere (presumably in countries where the tax burden is lower) and would end up being overcharged if they paid more in the UK. It should also not be forgotten that they generate VAT and pay income tax and national insurance for their employees, but nonetheless their tax affairs are more akin to the affairs in a bordello than those of honour.

spectrum pricingBut it seems that such activities are not just connected with foreign companies. Arqiva, the company responsible for transmitting all terrestrial television programmes across the UK (and many of the satellite programmes too) has, according to the Sunday Times, also been using some creative accounting to reduce its tax bill. The 'fiddle' they have employed is to get their parent companies (who are shareholders) to loan them money, but at a high interest rate, far above that which they would need to pay to take a loan at the bank. The repayments for these loans are taken from the accounts before dividends or the profit for tax purposes are calculated thus providing excellent returns to shareholders and also reducing Arqiva's corporation tax liability.

Arqiva's accounts show that shareholders took out 120m in 2012 and 106m in 2011 on loan interest charged at the stratospheric interest rate of 18%. Not quite Wonga.com interest rates but higher even than interest rates charged by car loan companies. As with Starbucks and Google, such fiddles are not illegal, but it could be argued that they are morally suspect, especially in the current financial climate. Arqiva are currently a monopoly provider and as such their fees are regulated. The regulated fees are based on their accounts, but such 'fiddles' would also impact the prices they charge. As their main customers include the BBC, UK citizens' licence fees are ending up in Arqiva's shareholders' pockets.

Arqiva would no doubt argue that as they do not make profits, they would not pay any corporation tax anyway. But they would be much closer to profitability if they weren't paying their owners for a loan at such extreme interest rates. What is certain is that their owners are benefiting from a much higher return on their investment than licence payers are!
add comment ( 1909 views )   |  0 trackbacks   |  permalink   |   ( 3 / 972 )

<<First <Back | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | Next> Last>>