What causes the 56kbps limit on dial-up internet connections?

Hi:

What are the physical* reasons dial-up speeds cannot go above 56kbps? One factor is that the phone line cuts of frequencies below 300 Hz and above 3,000 Hz. What are the other physical causes of this limit? Would this limit still exist if it weren't for the aforementioned frequency-cutoffs?

bps = baud X number of bits per baud.

If only 1 baud is used, what is the maximum-bits-per-baud that can be used on a phone line without the frequency of the analog electric signal exceeding 3,000 Hz?

*NOTE: By "physical", I am referring to causes not associated with legal regulations - such as limitations place by the FCC and governments.

Thanks,

Radium

Reply to
glucegen1
Loading thread data ...

snipped-for-privacy@excite.com hath wroth:

The speed limit is the point where the error rate drops to the point where communications is futile within the audio bandwidth. The V.90 spec allows 56Kbits/sec download, but only 33.6Kbits/sec upload. (V.92 does 48Kbit/sec upload).

Yep. It would be 64Kbits/sec were it not for telco bit-robbing, where every 6th frame is stolen for signaling. The audio channel can handle the bandwidth, but when it hits the DS0 digital channels, bit robbing limits the bandwidth to 56K. Therefore, you only get: 8Kbits/sec * 7 bits = 56 Kbits/sec.

What do you mean by a "band"? Are you perhaps thinking of bits per baud?

You could go faster if you don't care about telco channel crosstalk or hitting the line protectors thus creating distortion.

You may find the following of interest:

Hmmm... It's 7 years old. Maybe I should update it.

Oh, it's you again.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
 Click to see the full signature
Reply to
Jeff Liebermann

"Jeff Liebermann"

** The limit is set by the restricted bandwidth AND the available to noise ratio of a telephone network voice circuit.

A guy named Shannon work out the maths, loooong ago.

It kinda like the inverse of the Nyquist sampling theorem.

formatting link

The formula for when the s/n ratio >> unity is:

Bits/sec = 0.33 x bandwidth (Hz) x s/n (in dB)

For a voice circuit that becomes:

0.33 x 3000 x 45 ( typical good line)

= 45,000 b/s

The actual s/n ratio on a given voice circuit, which includes analogue twisted wire lines and digital signals over co-ax or fibre varies - so

56K modems do an automatic line test to see just what capacity is available and adjust themselves to that.

...... Phil

Reply to
Phil Allison

"Phil Allison"

** My previous post was to:

" snipped-for-privacy@excite.com "

Whose post does not show on my server.

........ Phil

Reply to
Phil Allison

Okay.

What is the "0.33"?

If the max a voice circuit can take is 45 kbps, then how can dial-up speeds get above that?

Also if frequencies below 300 Hz and above 3,000 Hz are cut-off, wouldn't the bandwidth of a phone line be 2,700 Hz?

Reply to
glucegen1

snipped-for-privacy@excite.com hath wroth:

Look at a communications channel this way. You can get *ANY* speed through a bandwidth limited channel, up to a given error rate. If your application requires a very low error rate to function, then you have to have a good signal to noise ratio, and your thruput will be fairly small. However, if you have a mess of forward error correction, small packets, and a very tolerant application, you migtht be able to squeeze some more thruput through the same channel.

I'm not going to expound on how V.90 works in detail. It gets messy fast. There are modulation schemes for increasing the base 600 baud modulation rate (bits per baud) to much higher bits/baud. Then add, adaptive equalizers, echo cancellers, error detection, error correction, data compression, etc. Anything to squeeze more thruput into a rather ugly looking POTS line. However, that's just between the user and the CO (central office). Once at the CO, everything gets converted to digital and the rules change.

Nope. I don't want to get into how a DS0 (digital) line work. You can get 64Kbits/sec out of a DS0 if you can use out of band signalling. However, if you're using in band signalling, you're stuck with 56Kbits/sec. Even if the analog part of the puzzle can go faster than 56Kbits/sec, the digital part at the CO will limit the speed to

56Kbits/sec.

Sorry. There was a fly splattered on my monitor.

56Kbits/sec. The limit is NOT all from the analog part of the line. The analog modem glop gets converted to digital at the CO and that's limited to 56Kbits/sec. I could easily (well maybe not so easily) get more than 56Kbit/sec thruput going between my house and the CO, but the digital thruput at the switch will limit thruput to 56Kbits/sec.

The problem is worse when dealing with SLC (subscriber line concentrators) where the analog to digital conversion is done outside the CO, such as with Pair Gain. The best you can do with those is perhaps 28.8Kbit/sec, mostly because the digital audio filter cuts off at a much lower frequency than the filters at the CO. Since most of the energy is in the higher frequency part of the audio spectrum, the loss of the higher frequencies is fatal to higher speed modem operation.

You really should be asking in comp.dcom.modems.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
 Click to see the full signature
Reply to
Jeff Liebermann

The fact that all modern telephone exchanges digitize the voice signal to 8 bits at a sample rate of 8 KHz, which means that the digital paths within and between phone offices absolutely cannot transport more than 64 KBPS. Some overheads, nonlinearities, and error correction things resulted in the 56K download number. Upload is burdened by some additional issues.

If the phone system were still truly analog, the number of bps you could jam into 3 KHz of bandwidth would depend on the s/n ratio, per Shannon, and could in some cases be a lot higher.

John

Reply to
John Larkin

You see, the internet is like a bunch of pipes with water running through it, think of water as your information. Plumbing is old, Pipes are small, You can only get some much through those pipes! It's true! One of our Senators even said so! :)

--
"I\'m never wrong, once i thought i was, but was mistaken"
Real Programmers Do things like this.
 Click to see the full signature
Reply to
Jamie

In reality I have rarely ever reached >30K in situations where I had to use dial-up. Hotels and such.

Lately our phone line here has a constant and very audible noise hash on them. Maybe they don't want folks using dial-up and buy the DSL service ;-)

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

The phone lines are digitised nowadays, the rate at wich they digitise them is I gues about 56kbps therefore no encoding scheme is going to get round this. the modems carefully tune the signal levels for each encoded state so that it matches the digitising of the phone line to as close to 1:1 as possible.

Colin =^.^=

Reply to
colin

"Jeff Liebermann"

** Bollocks.

You need to read up on Shannon's theorem - fool.

** Irrelevant to the inherent bit rate capacity of a voice communications line.
** Fly spots all over you as well.
** In truth - it is.

From the user's position a dial up voice circuit IS analogue.

....... Phil

Reply to
Phil Allison

snipped-for-privacy@excite.com snipped-for-privacy@excite.com posted to sci.electronics.design:

The number 1 reason is end equipment. If the end equipment does not support any better than POTS that is all you get. See also Shannon's theorem, which relates usable bandwidth, signal to noise ratio, and datarate. If, as usual, the end equipment is the limiting factor on a twisted pair line then that is the limit. If you are looking for the limits on just a Cat 3 copper twisted pair, then you are looking at something on the order of 30 MB/(second*mile) based on the equipment that i have seen for sale.

Reply to
JosephKK

There are a lot of old PBXes out there that really mess things up.

On good "direct line to the CO" connections, back when I used dial-up (>5 years ago now) I would routinely get ~45-48kbps. The only time I got the fabled 53kbps was with an internal (modern :-) ) PBX connection to an inside modem pool .

45-48kbps is really quite good given the transmission medium used, I think.

A lot of phone companies officially only support something like 14.4kbps and will do absolutely nothing to improve line quality if they can demonstrate a successful 14.4kbps connection. I can't imagine that part of that policy wasn't largely influenced by marketing... (Even so, of course, most lines will do much better.)

On the other hand, I'm not sure there ever was a means of determining that voice quality on a POTS line was "inadequate," was there?

I read an article the other day about some ham who was tracking down huge HF band interference at night. He quickly figured out that it was a bad ballast on a street lamp, but the power company made him go around and note all the bad lights he could find in something like a 2-3 mile radius. Supposedly they were "willing" to do this themselves, but the emissions/interference guys only worked regular 9-5 hours and were "not allowed" to work overtime... such as at night, when the street lamps would actually be *on*. What nonsense that is... I have to imagine there's a union in there somewhere... what hiring manager would ever hire someone to mitigate interference from streetlamps with the understanding that they'd never have to work outside of 9-5 unless overtime was paid?

---Joel

Reply to
Joel Kolstad

Theoretical as well as calculated limits may be nice to dream about, but the fact is that almost every telco throttles the speed to about

48K, making them liars when they say that they support 56K.
Reply to
Robert Baer

What telco says they support 56 kbit/s? I think you may be the liar.

Reply to
Don Bowey

Robert Baer hath wroth:

Everyone lies, but that's ok because nobody listens.

I routinely get 49.3Kbits/sec on my dialup connections. Sometimes, it goes to 50.3Kbits/sec. I guess the mythical throttle doesn't apply to me.

The actual maximum speed is about 53Kbits/sec. The FCC limits the line levels on dialup lines to prevent crosstalk. Without sufficient line level, the S/N ratio drops, which causes the error rate to climb. The modem compensates by reducing its speed.

Incidentally, the local telco only "supports" 1200 or 9600 (v.32) baud, depending on whom I ask. Just try calling 611 with a modem problem and see what happens.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
 Click to see the full signature
Reply to
Jeff Liebermann

After all, nearly all of them also offer DSL over that same line and they would really, really like that dial-up customer to fork over those additional monthly fees.

You can almost bet on that.

One reason why I have never worked in any larger company. Of course that didn't help us when we got that dreaded new "anything over 8hrs/day must be paid as overtime" rule in CA. Before, we could do 4/10 shifts which really helped families with sick relatives or lots of kids plus it helped us maintain a 4hr PM window. This stupidity has cost jobs. It also destroyed home-care for some families. I had people in my office in tears. I guess the unions thought they won but the reality is that the people they were supposed to protect lost. Much of this work is now outsourced. What choice do you have if they press you out of the profitability margins?

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.