home *** CD-ROM | disk | FTP | other *** search
- Path: turnpike.com!DavidE
- From: Dave English <DavidE@turnpike.com>
- Newsgroups: comp.dcom.modems
- Subject: are 'byte-wide' modems possible - yes we are all using them
- Date: Fri, 12 Apr 1996 12:00:22 +0100
- Organization: Turnpike Ltd
- Distribution: world
- Message-ID: <XSOsHhAGfjbxoAf8@turnpike.com>
- References: <Ken.Crossman-1104961348100001@ts1-port4.mas.ualberta.ca>
- NNTP-Posting-Host: pillar.turnpike.com
- X-NNTP-Posting-Host: pillar.turnpike.com
- MIME-Version: 1.0
- X-Newsreader: Turnpike Version 1.12 <U2yaxlNz9m7tpk5wwwfqeW1so7>
-
- In article <Ken.Crossman-1104961348100001@ts1-port4.mas.ualberta.ca>,
- Ken Crossman <Ken.Crossman@ualberta.ca> writes
- >The latest word about modems has it that 33.6kb is about the limit. If I
- >understand modem transmission correctly, the modem encodes data onto a
- >phase-shifted audio carrier, with the speed of transmission governed
- >partly by the number of possible phase shifts per cycle. Could a modem be
- >designed such that each phase shift represents a complete 8-bit word, ie.
- >if there were 256 possible discrete phase angles then 256 (or 2**8) unique
- >digital words could be represented? The transmitting modem would encode
- >the 8-bit word received from the computer to one of the 256 possible phase
- >angles. The receiving modem would translate the phase angle back into the
- >8-bit word it represents. Or is this 'byte-wide' technology now in use?
-
- As you say, something essentially like that, only rather more
- sophisticated, is already in use.
-
- Here is a brief explanation I wrote recently for another purpose. This
- is on the boundary of my knowledge, so others will have to correct me if
- I should step out of the truth circle.
-
- In V.34 the modems probe the line quality between them, in order to
- determine how their telephone connection behaves. The probe includes
- frequencies from 150Hz to 3750Hz, if you listen then you can hear them,
- they are the characteristic sound of V.34 as against V.22bis. The
- modems do this first in order to determine what corrections they can
- apply for defects in the line. Secondly they determine what the line is
- capable of in terms of transferring information. Knowing this they then
- decide upon the details of the method by which they will communicate he
- users data bits over the telephone connection.
-
- The method used is always to send symbols at a fixed rate. These
- symbols are actually two dimensional points within a signal
- constellation. The modem has a number of symbol rates that it can
- choose between and a number of possible constellations, each a sub-set
- of a master 960 point super-constellation. The raw rate at which the
- modem can transfer data is then the product of the rate at which symbols
- are transferred and the amount of information in each symbol.
-
- The possible symbol rates are:
-
- 2400, 2743, 2800, 3000, 3200 and 3429 symbols per second
-
- 2400 sps can support data rates from 2400 to 21600 bps
- 2743 sps can support data rates from 4800 to 24000 bps
- 2480 sps can support data rates from 4800 to 24000 bps
- 3000 sps can support data rates from 4800 to 26400 bps
- 3200 sps can support data rates from 4800 to 28800 bps
- 3249 sps can support data rates from 4800 to 28800 bps
-
- each depending on which constellation is chosen.
-
- During a connection I think that the modems can negotiate a change in
- the constellation used, according to experience of the success or
- failure to transfer data correctly. Only by doing a retrain can the
- modems change the symbol rate during a connection.
-
- Some modems can now communicate at up to 33600 bits per second using an
- extension of the V.34 standard, being called V.34+ . I believe this
- just involves using the highest information content constellations with
- the highest symbol rate(s).
-
- Regards
- --
- Dave English, T U R N P I K E Ltd
- Dorking Business Park, DORKING, Surrey, UK. RH4 1HN
- My other computer's a brain
-