Well the broadband dongle worked..but..

Just thought I would report back on the use of the three network USB
dongle when I finally got the Linux laptop to an area where there was
internet broadband mobile coverage.

Basically it did work. But for some reason the DNS did not. Three
offered the right DNS servers, but despite use of the 'usepeerdens' in
what I think were the right places, it refused to use them. Fortunately
I discovered it was running BIND anyway, and do just told it to use
localhost, whatever.

(Linux/debian lenny ppp etc etc)

The other odd thing was when I arrived on site to set it up, late
afternoon, we got good throughput..I didn't check, but it seemed about a
couple of hundred kbps.

Later on, at dusk, we seemed to be on 'crap dialup speeds' - looked like
a 9600 modem!

Finally, after dusk,we got no connection at all. A friend with a 3G
phone saw exactly the same degradation..

Can anyone explain why this is?

In article <gtgrpm$j7t$(E-Mail Removed)>, The Natural Philosopher
<(E-Mail Removed)> scribeth thus
>Just thought I would report back on the use of the three network USB
>dongle when I finally got the Linux laptop to an area where there was
>internet broadband mobile coverage.
>
>Basically it did work. But for some reason the DNS did not. Three
>offered the right DNS servers, but despite use of the 'usepeerdens' in
>what I think were the right places, it refused to use them. Fortunately
>I discovered it was running BIND anyway, and do just told it to use
>localhost, whatever.
>
>(Linux/debian lenny ppp etc etc)
>
>The other odd thing was when I arrived on site to set it up, late
>afternoon, we got good throughput..I didn't check, but it seemed about a
>couple of hundred kbps.
>
>Later on, at dusk, we seemed to be on 'crap dialup speeds' - looked like
>a 9600 modem!
>
>Finally, after dusk,we got no connection at all. A friend with a 3G
>phone saw exactly the same degradation..
>
>Can anyone explain why this is?

Kids home from school;?...
--
Tony Sayer

On Sat, 02 May 2009 08:12:52 +0100
The Natural Philosopher <(E-Mail Removed)> wrote:

> The other odd thing was when I arrived on site to set it up, late
> afternoon, we got good throughput..I didn't check, but it seemed about a
> couple of hundred kbps.
>
> Later on, at dusk, we seemed to be on 'crap dialup speeds' - looked like
> a 9600 modem!
>
> Finally, after dusk,we got no connection at all. A friend with a 3G
> phone saw exactly the same degradation..

3G W-CDMA suffers from a phenomenon called "cell breathing". This
occurs because unlike GSM where each logical link has its own timeslot,
in 3G all the signals overlay each other on a single frequency but the
different spreading codes allocated to each separate them in a similar
way. The link power is accurately controlled so that all the signals
arriving at the base station are at the same level (within 0.2dB I
think) to ensure that no one signal can overload the analogue to digital
converters in the receiver.

If there are many signals, the effective sensitivity of the receiver
reduces a little; think of it as the wanted-to-interferer power ratio
decreasing, the bit error rate goes up a bit. So, if you are at a fair
distance from the node, you get better performance when there are few
other users but as the user numbers increase things get worse. Taken to
its logical conclusion, it may well cause you to be unable to connect
at all if you are at just beyond the minimum cell radius for the path
loss between you and the node.

Take the above with a small pinch of salt, you may be seeing a different
effect, but the one I have described definitely exists.

--

Brian Morrison

bdm at fenrir dot org dot uk

"Arguing with an engineer is like wrestling with a pig in the mud;
after a while you realize you are muddy and the pig is enjoying it."

GnuPG key ID DE32E5C5 - http://wwwkeys.uk.pgp.net/pgpnet/wwwkeys.html

Brian Morrison wrote:
> On Sat, 02 May 2009 08:12:52 +0100
> The Natural Philosopher <(E-Mail Removed)> wrote:
>
>> The other odd thing was when I arrived on site to set it up, late
>> afternoon, we got good throughput..I didn't check, but it seemed about a
>> couple of hundred kbps.
>>
>> Later on, at dusk, we seemed to be on 'crap dialup speeds' - looked like
>> a 9600 modem!
>>
>> Finally, after dusk,we got no connection at all. A friend with a 3G
>> phone saw exactly the same degradation..
>
> 3G W-CDMA suffers from a phenomenon called "cell breathing". This
> occurs because unlike GSM where each logical link has its own timeslot,
> in 3G all the signals overlay each other on a single frequency but the
> different spreading codes allocated to each separate them in a similar
> way. The link power is accurately controlled so that all the signals
> arriving at the base station are at the same level (within 0.2dB I
> think) to ensure that no one signal can overload the analogue to digital
> converters in the receiver.
>
> If there are many signals, the effective sensitivity of the receiver
> reduces a little; think of it as the wanted-to-interferer power ratio
> decreasing, the bit error rate goes up a bit. So, if you are at a fair
> distance from the node, you get better performance when there are few
> other users but as the user numbers increase things get worse. Taken to
> its logical conclusion, it may well cause you to be unable to connect
> at all if you are at just beyond the minimum cell radius for the path
> loss between you and the node.
>
> Take the above with a small pinch of salt, you may be seeing a different
> effect, but the one I have described definitely exists.
>
Ah..so as people start turning up, with phones, that's likely to bugger
the connection.The reduced S/N in the presence of many spread spectrums
devices makes PERFECT sense, thanks.

We were somewhere around Barstow, on the edge of the desert, when the
drugs began to take hold. I remember The Natural Philosopher
<(E-Mail Removed)> saying something like:

>> Take the above with a small pinch of salt, you may be seeing a different
>> effect, but the one I have described definitely exists.
>>
>Ah..so as people start turning up, with phones, that's likely to bugger
>the connection.The reduced S/N in the presence of many spread spectrums
>devices makes PERFECT sense, thanks.

Also, if you are right at the edge of the cell coverage, foliage and
atmospherics play a part in killing (or helping) the signal.

Grimly Curmudgeon wrote:
> We were somewhere around Barstow, on the edge of the desert, when the
> drugs began to take hold. I remember The Natural Philosopher
> <(E-Mail Removed)> saying something like:
>
>>> Take the above with a small pinch of salt, you may be seeing a different
>>> effect, but the one I have described definitely exists.
>>>
>> Ah..so as people start turning up, with phones, that's likely to bugger
>> the connection.The reduced S/N in the presence of many spread spectrums
>> devices makes PERFECT sense, thanks.
>
> Also, if you are right at the edge of the cell coverage, foliage and
> atmospherics play a part in killing (or helping) the signal.

Not much foliage in an airfield..

In article <(E-Mail Removed)>, Grimly
Curmudgeon <(E-Mail Removed)> scribeth thus
>We were somewhere around Barstow, on the edge of the desert, when the
>drugs began to take hold. I remember The Natural Philosopher
><(E-Mail Removed)> saying something like:
>
>>> Take the above with a small pinch of salt, you may be seeing a different
>>> effect, but the one I have described definitely exists.
>>>
>>Ah..so as people start turning up, with phones, that's likely to bugger
>>the connection.The reduced S/N in the presence of many spread spectrums
>>devices makes PERFECT sense, thanks.
>
>Also, if you are right at the edge of the cell coverage, foliage and
>atmospherics play a part in killing (or helping) the signal.

Yes we're some 900 metres from a 3 and Vodafone main site. 3 is almost
useless indoors and not to good outdoors and Voodo fone is marginal
inside and OK outside but rarely more then a bar of level on the meter.

And all that gets worse in summer;(...
--
Tony Sayer

On Sat, 02 May 2009 13:50:54 +0100, The Natural Philosopher
<(E-Mail Removed)> wrote:

>Not much foliage in an airfield..

You've not been to Ed's then - 80 foot hedge an 120 foot trees...

--
Geo

In article <(E-Mail Removed)> , Brian
Morrison <(E-Mail Removed)> scribeth thus
>On Sat, 02 May 2009 08:12:52 +0100
>The Natural Philosopher <(E-Mail Removed)> wrote:
>
>> The other odd thing was when I arrived on site to set it up, late
>> afternoon, we got good throughput..I didn't check, but it seemed about a
>> couple of hundred kbps.
>>
>> Later on, at dusk, we seemed to be on 'crap dialup speeds' - looked like
>> a 9600 modem!
>>
>> Finally, after dusk,we got no connection at all. A friend with a 3G
>> phone saw exactly the same degradation..
>
>3G W-CDMA suffers from a phenomenon called "cell breathing". This
>occurs because unlike GSM where each logical link has its own timeslot,
>in 3G all the signals overlay each other on a single frequency but the
>different spreading codes allocated to each separate them in a similar
>way. The link power is accurately controlled so that all the signals
>arriving at the base station are at the same level (within 0.2dB I
>think) to ensure that no one signal can overload the analogue to digital
>converters in the receiver.
>

Thats interesting how they see to it that it does that!..

>If there are many signals, the effective sensitivity of the receiver
>reduces a little; think of it as the wanted-to-interferer power ratio
>decreasing, the bit error rate goes up a bit. So, if you are at a fair
>distance from the node, you get better performance when there are few
>other users but as the user numbers increase things get worse. Taken to
>its logical conclusion, it may well cause you to be unable to connect
>at all if you are at just beyond the minimum cell radius for the path
>loss between you and the node.
>
>Take the above with a small pinch of salt, you may be seeing a different
>effect, but the one I have described definitely exists.
>
So why would this get worse as the day goes on when the cell traffic
reduces , well mobile voice does a bit tho the net is possibly likely to
go up...

And thats all with a pinch of snuff..
--
Tony Sayer

On 2 May, 14:55, tony sayer <t...@bancom.co.uk> wrote:
> In article <20090502114900.70cf6...@peterson.fenrir.org.uk> , Brian
> Morrison <b...@fenrir.org.uk> scribeth thus
>
>
>
> >On Sat, 02 May 2009 08:12:52 +0100
> >The Natural Philosopher <t...@invalid.invalid> wrote:
>
> >> The other odd thing was when I arrived on site to set it up, late
> >> afternoon, we got good throughput..I didn't check, but it seemed about a
> >> couple of hundred kbps.
>
> >> Later on, at dusk, we seemed to be on 'crap dialup speeds' - looked like
> >> a 9600 modem!
>
> >> Finally, after dusk,we got no connection at all. A friend with a 3G
> >> phone saw exactly the same degradation..

Might be worth coinsidering getting a USB extension cable and
hanging your important bits out the window - or otherwise
positioning them advantageously.

USB cables are limited in length.

I read recently of wireless routers that would also
take 3G USB. Could then stick that well away from the
computer and use 802.11 or ethernet or combination thereof
for the link to it.


http://www.usbwifi.orconhosting.net.nz/
would work too I would imagine.