BBC News Blunder

On 31/07/2017 10:45, Brian Gaff wrote:
Yes you can position them anywhere, but the point is that it has to be at
the equator.
Obviously if the position was so low in the sky from the uk that it was
occulted by the land or buildings, its not going to work well and you would
be unlikely to be in its coverage area for any of the beams sent to earth or
receiving from it unless the programme material was uplinked from here.
Brian

There's also a problem with atmospheric absorption when the satellite is
only just above the horizon.

Anyone remember the Reykjavik Summit?

Bill

On 31/07/2017 10:58, Brian Gaff wrote:

That is a long way out and why the footprints of coverage are large.
Brian


You 'can be large'.

Bill

On 31/07/2017 11:56, NY wrote:

An orbit parallel to the equator but at a latitude of (for example) 50
degrees (for northern Europe) will require constant burning of fuel to
keep it overhead, so it's not viable.

That wouldn't be an orbit' it would be a flight path.

Bill

Jim Lesurf wrote:

In article , 7
wrote:

No because Earth is a slight doughnut shape with the bulge through the
equator, and the Moon is also orbiting around the equator which means
orbits other than equator are not stationary or stable.

IIUC In reality, none of the usual orbits are absolutely stable. Just that
some get closer than others. Even geostationary satellites tend to require
'stationkeeping' thrust burns and may drift away from their nominal
location when that is no longer possible. The usual circles and ellipses
are just a first approximation to real behaviour.

IIRC it is routine to 'park' geostationary sats elsewhere as they near the
end of the design life.

Hmm... None of the other trolls like Bill Wright
understood their satellite subject matter it seems.

They clean forgot to mention the +/-2cm deviation in estimates subject
to measurement uncertainty.

Tut tut.

In article ,
NY wrote:
"Brian Gaff" wrote in message
news
I often wonder aboutsat broadband though as surely it will be painfully
slow with every packet taking significant time to go up and down again.

[ snips ]

I've forgotten what the maximum amount of data is that can be sent without
acknowledgement, but I seem to remember it's only a mater of a few 1500-byte
packets, so that means that every few kB of data needs a 0.25 second pause
while the sender says "OK. Ready for some more data" and another 0.25 sec
before the sender's data arrives. This is what slows down the overall data
transfer.

It's been some years since there was such a severe underlying limitation
- RFC1323 is dated May 1992. For the details read up on TCP Window
Scaling, Bandwidth Delay Product and Long Fat Networks.

Nick
--
"The Internet, a sort of ersatz counterfeit of real life"
-- Janet Street-Porter, BBC2, 19th March 1996

This topic was raised this morning, in LBC Nick Ferrari spot.

He proudly said he had no idea what the speeds really meant (which I can
quite believe!), and so he invited people to phone in and tell him how
many times faster 28Mb/s was compared with 500kb/s.

I think it took six callers to get the right answer - but even he said
that at first he was going to say "17". But like many who should know
better, he kept referring to 'megabytes' instead of 'megabits per
second'.

Doh!


In message , Java Jive
writes
"Powys man gets broadband via satellite over Africa"

http://www.bbc.co.uk/news/uk-wales-mid-wales-40745533

"[It is the same satellite companies like Sky use to broadcast their
TV signal and] they are positioned above the equator to maximise
coverage."

Oh dear! Actually, they are positioned over the equator because that
is the only place you can obtain a geo-stationary orbit, enabling
fixed dishes to work. Satellites that aren't in geo-stationary orbit
require movable dishes that can track an orbit to communicate with
them ...

http://www.macfh.co.uk/JavaJive/Audi...SatelliteAnaly
sisClarke.html

--
Uan

"Alan White" wrote in message
...
GPS sats are not in all sorts of inclinations.

Here's the post I sent yesterday:-

quote
The GPS satellites are in 55 degree, semi-sidereal orbits. This means
that their orbits are inclined at 55 degrees to the equator and that the
orbital period is 11h 58m. The last time I checked there were 27 active
satellites. Reception from three satellites is required for a 2D fix,
four satellites for a 3D fix. The horizontal position will have a
typical accuracy of better than 15 metres anywhere on the globe for 95%
of the time over a 30 day period. The 3D position altitude error is
roughly twice the horizontal error.
/quote

Ah, I didn't know that all GPS satellites are in orbits at the *same* same
angle. I presume "55 degrees" means one direction, not two (ie + or - 55 deg
wrt equator), or they could collide!

"Ian Jackson" wrote in message
...
This topic was raised this morning, in LBC Nick Ferrari spot.

He proudly said he had no idea what the speeds really meant (which I can
quite believe!), and so he invited people to phone in and tell him how
many times faster 28Mb/s was compared with 500kb/s.

I think it took six callers to get the right answer - but even he said
that at first he was going to say "17". But like many who should know
better, he kept referring to 'megabytes' instead of 'megabits per second'.

I can understand why non-computer people refer to megabytes when they mean
megabits, because most other measures in computing are multiples of a byte
(eg a 20 kB document, 3 MB image file, a 500 GB or 1 TB hard disk drive
etc). Comms is different: speeds are usually quoted in multiples of *bits*
per second. Maybe they shouldn't be, and we should be quoting payload speeds
in multiples of bytes, excluding any housekeeping and error-correction bits.
I tend to work on a very approximate factor of 10 when converting MB/sec and
Mb/sec, allowing for 8 bits/byte plus a bit of housekeeping.

And also speeds should as far as possible always be quoted in the same
multiple - always in kb/sec or always in Mb/sec, not a mixture of the two.
Normal engineering practice is to use the largest power that is a multiple
of 1000 such that the significant digits are in the range 1-999 (*)

so 5 Mb/s or 28 Mb/s or 157 Mb/s, but 500 kb/s rather than 0.5 Mbps

But in this case, to make comparison easier, I'd say that this is a special
case and it *should* be quoted as 0.5 Mps.


(*) Hence it is far more logical for rev counters to be calibrated 0, 1, 2,
...., 7 x 1000 rpm, as they are now, rather than the old style of 0, 10, 20,
...., 70 x 100 rpm, because the latter multiplier is not a power of 1000. (It
was also prone to people getting the speedo and rev counter confused because
both were calibrated with numbers in the same range). And also, British
English references to "seven thousand five hundred" are more logical than US
English references to "seventy-five hundred" :-) I tend to use "hundred"
notation up to 2000, and "thousand" notation after that "fifteen hundred"
but "two thousand five hundred".

On Mon, 31 Jul 2017 13:48:32 +0100, "NY" wrote:

"Alan White" wrote in message
.. .
GPS sats are not in all sorts of inclinations.

Here's the post I sent yesterday:-

quote
The GPS satellites are in 55 degree, semi-sidereal orbits. This means
that their orbits are inclined at 55 degrees to the equator and that the
orbital period is 11h 58m. The last time I checked there were 27 active
satellites. Reception from three satellites is required for a 2D fix,
four satellites for a 3D fix. The horizontal position will have a
typical accuracy of better than 15 metres anywhere on the globe for 95%
of the time over a 30 day period. The 3D position altitude error is
roughly twice the horizontal error.
/quote

Ah, I didn't know that all GPS satellites are in orbits at the *same* same
angle. I presume "55 degrees" means one direction, not two (ie + or - 55 deg
wrt equator), or they could collide!

Wikipedia has an animated image showing GPS satellites orbiting the
Earth and the Earth rotating beneath them. It shows which satellites are
in view of a particular point on the Earth at any time. The satellites
in view are shown in red. A satellite not in view is black.
https://en.wikipedia.org/wiki/Global...#Space_segment

Caption:

A visual example of a 24 satellite GPS constellation in motion with
the earth rotating. Notice how the number of satellites in view from
a given point on the earth's surface, in this example in Golden,
Colorado, USA(39.7469° N, 105.2108° W), changes with time.

--
Peter Duncanson
(in uk.tech.digital-tv)

On Mon, 31 Jul 2017 10:55:26 +0100, "NY" wrote:

"Brian Gaff" wrote in message
news
I often wonder aboutsat broadband though as surely it will be painfully
slow with every packet taking significant time to go up and down again.

The big problem with satellite broadband is flow control. When you request a
web page or a download, it will indeed take a significant time for the
request to reach the server and for the returned data to start arriving.
https://en.wikipedia.org/wiki/Geostationary_orbit says that the satellite is
about 36,000 km "above the earth's equator" (I presume that's distance from
the surface, not the centre, of the earth.

Yes. That Wikip page says after an explanation of the orbit calculation:
"The resulting orbital radius is 42,164 kilometres (26,199 mi).
Subtracting the Earth's equatorial radius, 6,378 kilometres (3,963 mi),
gives the altitude of 35,786 kilometres (22,236 mi)."

So the transit time for the data will be 2 x 36,000,000 / 3x10^8 = 0.25
seconds.

Once you've started the flow of data (the initial pause) it will arrive as
quickly as it would over landlines.

But the receiving end needs to be able to control the flow of data, so it
doesn't arrive faster than can be processed. And then TCP flow-control
packets have to travel back by the same route and so will take that same
time to arrive at the sending end.

I've forgotten what the maximum amount of data is that can be sent without
acknowledgement, but I seem to remember it's only a mater of a few 1500-byte
packets, so that means that every few kB of data needs a 0.25 second pause
while the sender says "OK. Ready for some more data" and another 0.25 sec
before the sender's data arrives. This is what slows down the overall data
transfer.

--
Peter Duncanson
(in uk.tech.digital-tv)