🚀 go-pugleaf

In article <Jim_Sackman.18.001026A1@usa.racal.com>,
Jim Sackman <Jim_Sackman@usa.racal.com> wrote:
>In article <NEWTNews.820626346.32539.jrhodes@jrhodespc.nwest.mccaw.com> jrhodes@jrhodespc.nwest.mccaw.com writes:
>Well, one reason for 56K B channels beyond what Fred said (which are 
>technically correct) is that PACBELL uses its Switched 56 network to run ISDN 
>data calls.  An excessive all 0's is not a condition to start a Yellow alarm, 
>unless it causes a Red condition.  LOS (Loss of Signal) would occur first.

   But if the question is "when do ISDN VOICE calls not get 64K unmodified?"
then PacBell's Switched-56 network wouldn't enter into the answer.  As far as
I know, they are keeping VOICE and DATA traffic segregated by trunk group.
If they eliminate any remaining non-SS7 Switched-56 trunks, then they could
merge voice and 56K over the same trunks.  Eventually, with B8ZS and/or
DS3/SONET switch interfaces, they could also support 64K data over those
same trunks (as some TELCOs already do).

>Also, even though the 8th bit is used for signalling out of every 6th frame, 
>you should not count on the other 5.  Depending on how your network is 
>provisioned, the carrier may have to terminate the T-1 (like at a 1/0 DACS).  
>This can cause "Signalling re-insertion".  This is used when a T-1 superframe 
>is torn apart to re-align the signalling bits in the superframe.

   The signaling bits are also re-arranged every time the call goes through
a tandem switch.  Some ITU-T document (G.714?) actually indicates the
probability of 1, 2, ... 6 bits of every 6 being forced to 1 by tandems on
an International call, and the resulting degradation in S/N ratio.

Al Varney