Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!cs.utexas.edu!tut.cis.ohio-state.edu!mailrus!accuvax.nwu.edu!nucsrl!telecom-request From: steele@ee.ecn.purdue.edu (Richard Steele) Newsgroups: comp.dcom.telecom Subject: Re: Data Over Voice Message-ID: <2211@accuvax.nwu.edu> Date: 15 Dec 89 09:16:18 GMT Sender: news@accuvax.nwu.edu Reply-To: Richard Steele Organization: Purdue University Engineering Computer Network Lines: 105 Approved: Telecom@eecs.nwu.edu X-Submissions-To: telecom@eecs.nwu.edu X-Administrivia-To: telecom-request@eecs.nwu.edu X-Telecom-Digest: Volume 9, Issue 577, message 5 of 5 Thanks to everyone who replied (both by mail and news) to my query about "data over voice" (DOV) units. I'd like to reply to everyone who responded, but since it's finals week here (I'm sure everyone remembers those), I'll save time by responding to [Michael.Fryd@g.gp.cs.cmu.edu] since his experience with DOV is very close to mine: >Around here, DOV means "Data Over Voice" and refers to the Gandlaf DOV >640 modems that provide 64kb synchronous communications over a >standard voice pair, without interfering with normal voice service. This is exactly the unit we have here. >In Pittsburgh, this service is known as Metropolitan Campus Network >(MCN) and is provided by some combination of Carnegie-Mellon >University (CMU) and Bell of PA. This is similar to the arrangement Purdue has... >Two DOV modems are actually used for each house served. One is placed >at the Bell of PA Central office(CO), between the switch and the cable >pair to your house. The second DOV is installed in your house, >electrically between the Network Interface Jack and your existing >household phone wires. >+---------+ >| Central | voice +--------+ data and voice +---------+ voice +---------+ >| Office |-------| CO DOV | ----------------| user DOV|-------|telephone| >| switch | +--------+ +---------+ +---------+ >+---------+ | | > |data |data >Port Selector in CO--- + +-------Home terminal or PC >The modems communicate using frequencies above the voice band. The >DOVs completely ignore all of the normal telephone signals, allowing >uninterrupted data communications regardless of whether the phone line >is in-use, ringing, or idle. The DOV operates as 64Kb synchronous. >In order for the DOVs to work you must be within a pre-set distance (I >think on the order of a few miles) from the CO. I believe this was the general concensus of the many replies I've gotten. Perhaps my biggest problem was that I didn't realize the simple twisted pair could handle such large bandwidths. Why is there a limit on the distance to the CO? Is there any reasonable way to increase this? >The real problem with this setup, is that it only allows you to communicate >with the CO. Bell of PA installed port selectors in three of the COs >near CMU. These were connected by fiber to CMU. [more info deleted] This is where the Purdue system differes (slightly) from the one at CMU. Since Purdue effectively owns the local network (i.e. we have to dial "9" to get an off-campus number), the "CO" is actually on-campus, very near the computer setups. Our hookup is to an ISN (Integrated Services Network), which is, I believe, a product from AT&T. We can go from there directly to most campus networks or go to the SDS (Serial Data Switch) to get to others, including those off campus. (I could, if I had the desire and the account, log onto a computer at Indiana University from the comfort of my own home! Life _is_ wonderful...) >When I left the program, only 9600 and 19.2 kBd async was supported, >with plans for 64Kb, possibly using Serial Line IP. Unfortunately, ISN (I think as Purdue has it configured) only supports 9600 baud. Unless I'm doing a _big_ file transfer to my home PC, this is very acceptable. Of course, there's the old programming adage that a project will fill the time alotted; a loose corollary would be a user always fills the bandwidth provided... :-) :-) >The most amazing thing was that the system worked very well. Voice >and data operated independently on the same copper pair. Incoming and >outgoing calls did not affect data at all. I definately agree. The system is _much_ slicker than using a 2400 baud dialup. Not only is it faster, but my roomates are much happier now that they have a phone again. >I have no idea what the applicable tariffs were. Payments were made >to CMU, but installations and service was provided by Bell of PA. I >was always amazed that the Bell service operators (just dial 611) were >able to cope with MCN trouble reports. We make our payments to our residence hall; the program right now is limited to most of the housing units, with plans to expand to off-campus users. I'm not sure how this would affect the CO connections, since Purdue has its own CO -- I hate to think of the mess it might cause! My roomate (_he's_ the expert in telephony) insits we should just go to ISDN; I'll bring him back to this century sometime before lunch. :-) As for cost, we pay $100 for the academic year, which includes charges from the CO. The unit is rented and there is no monthly fee. From several responses I've gotten, I'd say this is a competetive deal... >Michael Fryd >President Voice: (412) 751-5557 >MEFCO, Inc. Fax: (412) 751-8403 >2401 Coulter Road Email: Michael.Fryd@CS.CMU.EDU >McKeesport, PA 15131-4251 Thanks for the helpful response... Richard A. Steele Purdue University