|
The Digital Future Of The Telephone Network
A Study of Evolving Technology
By Lee Goeller
Originally published by Probe Research Inc. 1979.
Reprinted by permission
Chapter 9
Some Possible Shortcuts
to the Digital Future
If, indeed, 60% of all Bell System customers
are going to be served by analog ESS switches by 1985 and, further,
if these switches turn out to have the 40 year life that telephone
equipment has traditionally been designed to provide, the
all-digital future is a long way off. Still, it is possible to hope.
The facts to keep in mind are these: Digital
data is needed primarily by large businesses; large businesses are
almost all PBX or Centrex customers; business customers tend to be
localized in business or industrial areas; large business customers
have only 14% of the telephones in the country; this limited number
of highly concentrated telephones generates most of the toll traffic
during business hours and a great deal of the local traffic; the
needs of business customers are different from those of residential
customers.
In metropolitan areas (areas served by
several telephone central offices), a relatively small number of
calls terminate in the same office in which they originate. With the
100,000 line No. 1A ESS coming along, this will change slightly, but
even the 1A will do little about toll calls. Business lines generate
more calls, and a higher proportion of toll calls, than do
residential lines, and PBX trunks to a central office have very high
occupancies. The obvious suggestion is to use digital PBXs and home
them directly on No. 4 ESS tandem/toll switches. This would permit
digital service to be extended to the customers who need it and it
would prevent the waste of analog Class 5 offices which, at present,
simply slow down the flow of business traffic, most of which is
bound elsewhere. Residential and small business customers, whose
needs and traffic patterns match well the current crop of ESSs,
would be adequately served and the Bell System would have some time
to get back its investment on existing equipment.
The ideal PBX, then, would use T-compatible
digital switching, and would use digital trunks to the No. 4 ESS. It
should be capable of handling CCIS signaling to speed DID and to
minimize post-dialing delay, and it should maintain 4-wire, if not
full digital, integrity to the station set to minimize any possible
transmission problems. Such a PBX should permit direct digital
access to the bit-stream, whether at the subset, the remote
concentrator, or the distribution matrix. For point-to-point
(non-switched) lines, the matrix should be capable of providing
patch-panel operation without losing traffic handling capability.
For such lines, a sub-multiplex terminal might be useful to put many
lower speed data streams on one 64 kb/s voice channel.
The No. 4 ESS would need little modification;
about the only new requirement would be a few per-line
sub-multiplexers and time-slot interchangers to permit unscrambling
lower speed data streams on a voice channel and sending them off to
DDS on a patch-panel basis. True, there would be a larger number of
smaller trunk groups, and "fill" would be a problem since few PBXs
need CO trunks in groups of exactly 24, but a trunk group of 20
running at P.05 has an occupancy of better than 65%; this, coupled
with the savings produced by NOT filtering the calls through a Class
5 office, should justify direct access on a No. 4 machine. If
telephone-company operator access turns out to be a problem, perhaps
a few TOPS positions could be obtained from Northern Telecom. It
would be silly to have to continue to put all calls through a Class
5 office just to be able to pick up TSPS on the few that require an
operator.
Note that regular, conventional PBXs can
access the all-digital network just like modern digital PBXs. All
AT&T would have to do is move the channel bank that does the A/D
conversion at the No. 4 ESS out to the customer's premises. Voice
channels could be connected directly to PBX trunk circuits, and data
channels, in the manner of some Independent Industry T-carrier
terminals, could connect via different plug-in line cards to data
trunks. Even the fill problem would be under control since here
extra channels could be used by other customers fairly easily.
One thing that might require some change
might be the tariffs, particularly those relating to tie lines. At
present, carrying all tie lines to the customer's premises for
switching is usually less expensive than just carrying the access
lines required by a PBX for access to the customer's tie line
network. This is strange, since many more voice channels must be run
from the telco to the customer location in the former than the later
instance. If access-line costs could be made economical for the
customer as compared to freebie "end circuits," CCSA, using No. 4
ESS for switching, could become highly useful and economical.
Further, by offering "business day" circuits at a lower cost, tie
lines, normally unused at night, could be made available to
residential callers after business hours. This might even make
possible lower off-hour rates.
It is unfortunate that the Dimension is not a
digital switch. However, a "black box" could be devised to take the
PAM samples, already present, and convert them to PCM to permit
direct access to digital trunks. Or, if worst came to worst, the
"conventional PBX" approach outlined above could be used. If the tie
line switching of Feature Package 8 could be rendered as a service
and be performed in the No. 4 ESS, the best of several possible
worlds could be enjoyed.
Although relatively straightforward and easy,
even the above is not likely to happen. The anti-digital set of mind
in the Bell System seems to be totally uncompromising. The need for
a digital switched network simply is not considered very pressing.
"What have we missed by NOT being digital?" is the question Bell
people throw back at me. It is a hard question to answer. One can
argue by analogy, showing the well-documented loss to the telephone
industry produced by delaying the policy change that permitted
facsimile to get rolling, but it is hard to show how all the present
digital data, happily modernized, would be better off in digital
plant rather than analog. The fact that present private and
value-added data networks are adequate seems to be sufficient
justification for sticking with analog Class 5 switches, analog
PBXs, analog remote concentrators and analog station carrier.
The major clue to the Bell attitude is their
insistence that savings come from stored program control and not
digital transmission. The supposed savings that stored program
control produces are NOT in equipment, but in proposed new features
and services. These new features and services, however, are not for
the users, but for the telephone industry: better routing, network
administration, maintenance efficiency, etc. etc. Since stored
program control is used with digital switching as well as analog,
the obvious equipment savings that digital operation permits are
strangely underplayed. Thinking back, I remember vast* arguments
about the equipment savings possible with PNPN crosspoints vs. gas
tubes, or LLL logic vs. TRL logic. Extrapolations of hardware costs
into the future have a great track record for absurdity, but saving
people-costs just can't lose. Nevertheless, the analog option in the
face of LSI has to be one more step backwards.
[*Footnote:
Would you believe half?]
But this we know: the shape of future
communications is digital, and digital techniques that can be used
for both voice and data, adding the latter at minimal incremental
cost, can open up new opportunities and be available to meet present
needs more effectively. If we go in the right direction, we will
ultimately reach the right goal. If we are led off into
irrelevancies on the basis of pre-obsolete device costs which
underestimate the ingenuity of LSI developments and costs for a
per-telephone market, we may never get anywhere, and never even know
what we have missed.
In the words of the immortal Edison (who,
among other things, invented the carbon microphone still used in
your 500 type telephone set), "Nothing worthwhile ever works just to
please you. You have to make the damn thing work." There are
difficulties and problems on the way to the all-digital network. A
few specialized common carriers willing to innovate would be a big
help, but that is too much to hope for. If it's ever going to fly,
we, the customers, perhaps with some help from the Independent
Telephone Industry, will have to make ourselves heard. If we keep
pressing in the right direction, maybe some day the Bell System will
come around. Right now, the System is the Problem. But it doesn't
have to be that way. I cast one vote for the digital future.
[ Top ] [
Table of Contents ] [ Next Chapter ] |
