At 00:21 02/10/99 +1000, Paul B. Webster VK2BZC wrote:
>Gerhard Fiedler wrote:
>
>> yes, you're right, it's 16kHz in Germany as well. and it's gotten
>> slightly confusing there since I left: they opened the long distance
>> carrier market, and since the telcom switches don't transmit billing
>> information, calls using one of the optional carriers (a similar
>> access system as with the 1010257 or so prefixes here in the usa)
>> don't provide the billing pulse.
>
>  Of course they don't, because ... you don't want to be billed for
>them!  Billing pulses are to register charges to your primary (local)
>carrier.  If the others provided these pulses, you'd be paying double!

for one, that's apparently different in germany. the calls made through
these carriers get billed on the same main bill. and in any case, that's
not the issue: even if i got billed by them directly, they still could send
the local carrier the billing information (while i'm talking), and i could
get the pulses accordingly, without double billing (since the local carrier
would know that the billing information came from another carrier, and just
put the corresponding pulses on the line without charging me). it's just
that the switch protocols don't provide that information, i'm told. this
seemed to have been (or maybe still is) a big issue in germany, since a lot
of telco equipment and procedures relied on the pulses, and it seems to me
that if the switch protocol would make it possible to provide that
information, they probably would've designed something to get it done. but
i'm no telecom specialist...


>> many hotels and other enterprises seem to have troubles adapting to
>> the new situation of the possibility of having calls on the system
>> that don't provide a billing pulse but nevertheless get billed --
>> probably in the beginning a couple of verrry smart guys got away
>> without paying their phone bill :)
>
>  They would have to have a "lock" so that only one carrier is used.
>Administration otherwise would be too complex (though I reckon someone
>will have a go at this!).

yes, that's what they're doing now. it's just that for ages hotel systems
had all their billing (and billing equipment) based on the pulses, so this
change apparently had quite an impact.


>> a nice suggestion to wire a modem with a filter so that the modem
>> signal goes through the filter but not the phone signal (if you're
>> using a pulse counter in the phone).
>
>  16 kHz is so far from the upper line bandwidth (3 kHz or 3.3 kHz)
>that most modems *should* filter this automatically (DSP anti-alias).

i can tell you from my own experience that this is not true. at least
during the time i used imported modems in germany (that was pre-v.34, 10 to
8 years ago), most of them got heavy hiccups from the pulses (up to loosing
carrier without being able to reconnect -- the zyxel modems, with their
proprietary 19k2 protocol when everybody else was happy to get 14k4, were
particularily difficult), which could be solved by a filter -- that's where
those suggestions came from.


>> it has to be pretty hot, because the system wasn't designed for that
>> bandwidth, so the losses are not precisely calculable and there has to
>> be a pretty big margin.
>
>  Well, that's not quite so.  This signal is injected at the SLIC anyway
>as it is certainly not carried over the exchange circuits (or are we
>talking pre-digital systems?).

probably, at least historically. i think the billing pulse predates
all-digital systems; they may be it by now, but the standard probably has
been desgined to work on analog systems as well (or only).

>The subscriber line *can* handle that
>bandwidth because it at least potentially, can be used for higher-grade
>digital transmission.  Or so they're *hoping* for the next century.

is that true? i mean, i figure that =some= lines are good enough, but
that's not reliable. especially in countries like australia and the usa
with the long distances in some places, there seem to be a number of remote
subscriber lines which i assume would damp a 16kHz pulse quite
considerably. do you think you can go with, say, 12kbaud (not 12kbps! this
is of course possible) over all subscriber lines?


>> pretty lucky though that it doesn't actually get to the ears :))
>
>  Wouldn't be noticeable, since most adults are fairly deaf to it! When
>could *you* last hear the TV set's horizontal scan whine from the other
>end of the house?

not from the other end... but i uauslly don't hold the phone receiver at
the other end of the house, either :)

>It's most certainly filtered out by the telephone receiver.

i guess i'm in the age by now where my audio bandwidth is getting smaller;
i haven't tested it in a while but i'm still happy that the receiver isn't
a piezo or something which wakes up at 16kHz for real activity :)

ge