Broadband Over Copper Wires

Pretty well.

Idiot!

No -- it's even worse than that. At least idiots have an excuse. I know better, but am incorrigible. I'd make a crude stab at your question, but I think there are others who know this stuff inside out, and who would do a much better job.

Your question might be a little too broad, however. Maybe if you added a post giving more specifics, you'd get more responses. Here's a Wikipedia link that appears to give a good description, but there are others here who are better judges.

Thanks ken havn't had a laugh all day.

Laugh is highly contagious. Others should take care!

In its simplest form, it's the same as dialup using a modem, but with a special modem that's faster. Just because it's running on copper wires doesn't mean you can run a high bandwidth connection down it (just dont expect to get as high a data thru-put as you would a fibreoptic cable). Try doing a google search, there is plenty of info out there.

Ooooh, just remembered a story that was told to me yesterday by my flatmate (who works for the telephone company), our dialup broadband connection was running really slow for some reason. He ran our subscribers online broadband download speed test and it told him our download speed was 1MB in 0.07 seconds (which it also said was about 30MB/sec, <sigh>). He called up customer support and complained, and the Indian on the other end told him that the connection they provide was "very, very fast", flatmate told customer support he ran the download speed test and that it was grossly wrong as it was physically impossible to download that fast on our connection, customer support said yes "our connection, very, very fast". Customer support told him to load up the BBC homepage as a test. It took 25 seconds to load. Upon informing customer support of this he replied...........

..................................................................................

"Yes that is very, very fast. What is the problem?"

What is the problem indead?

Broadband communication is basically in 1MHz to 100MHz range frequency and it will change if you have larger distance of wire from service provider. Like a fax machine, handshaking can adjust the communication rate. Some boardband links are FSK narrow band and some may be spread spectrum. Other may be frequency hopping or frequency fadding.

Broadband means that your system can change the modulation frequency and data rate. This was not possible in many wireless systems as frequency is restricted. In copper, frequency does not occupy the radio bands in air as signal is on the copper. Some signal will sure getout as wire will act as antenna if it is not shielded.

You can also have broadband in power line and fiber optic communication.

When you say '...over copper wires...' I assume you are referring to coaxial cable and not just a twisted pair?

Providing coaxial cable is driven and terminated correctly with the characteristic impedance of the cable, you can get extremely high bandwidth through it...

High bandwidth = high speed = broadband operation...

John.

Dear John

On the contrary, you never know the frequency so you never know the exact impedance that is needed in broadband communication. It is a very funnt way to communication on unknown impedance transmission line.

This works this way. If I send something and you receive then you acknowledge else I change my band and look for you elsewhere. Once I find you, I can communicate.

Copper is talked about simple open telephone wires separated to get some 100 ohms to 300 ohms impedance. There is really no restriction on impedance and boardband is a technology and nothing to do specially with copper. That is just one way to do it.

God's message comes through lightning and that is real broadband I suppose. It also contains signature of the universe, was, is now and that it will be.

How are you going to celebrate this x-mas? Do you have enough ice there or is it warm?

Yes, I understand transmission line theory and practice Shyam.

A correctly terminated transmission line will, ideally, have a resistive impedance and thus not be frequency selective at all.

This ideal system is approximated by a coaxial line, using twisted pair or worse obviously the impedance would vary, making it a very unsuitable transmission medium.

Its fairly chilly here at the moment, although it has been a mild start to the winter, the temperature is now just about 0 *C.

As almost all my family have emigrated to the States many years ago I will be having a quiet Christmas... I think like most other holidays now that it is a highly commercialised holiday and I for one don't agree with the waste of money items which a lot of people seem to think they must have, or maybe its the advertiser's hype?

So I'm going to be having a quiet holiday time with some visitors etc...

How about you, I would have thought the climate in India would be excellent for a holiday at this time of year?

John.

My 2 Cents

I'll chime in on this for what its worth. The comment "copper wire" is highly misleading since the question could be referring to CAT 5e twisted pair or to telephone line installed in the '40s. Highly different configurations of "copper wire" with unpredictable impedance, etc.

I don't imagine the question refers to coaxial cable because most people with enough understanding to be questioning the bandwidth capability of "copper wire" would probably specify coax if that was their concern. Commonly you talk about coax in terms of lumped reactive and resistive constants.

If you are talking about a carrier with information encoded on it in some fashion, then you can probably obtain adequate impedance matching through common techniques to meet your needs but the PHYSICAL configuration of the "copper wire" will limit the upper freq of your carrier--remembering that this is not a brick-wall limit, but rather an arbitrarily-defined 3dB down point. Remembering also that we don't need to be thinking in terms of the old telephone 600 ohm power-transfer matching--all we really need to do is detect a binary change of state at the far end. As Shyam indicated, this can be done with spread spectrum techniques clear down in the noise floor.

We haven't even mentioned compression algorithms and error detection/correction techniques designed to get the information through even when we have radically exceeded the ordinary capabilities of "copper wire".

So the question I think, is a pretty good one. I wish I understood exactly what "Broadband over copper wire" really means.

Shyam, we use equipment (in radio broadcasting) that dynamically senses and adjusts to impedance and other line changes as we move digital audio over telephone infrastructure. Is this markedly different than what is done with modems to your knowledge? Have I said anything stupid in the foregoing? Anyone?

I certainly appreciate the knowledge I glean from these threads. Thank you all for you erudite input.

Lonnie

Hi Lonnie, you're possibly right... I'm not sure what the transmission modulation / demodulation signal of a broadband signal is?

All I do know is the trouble the telecom companies had in getting the modem speed up to 14k4 then 28k then 33k and finally up to 56k over the telephone lines...

I know they tried all sorts of weird and wonderful modulation and encoding algorithms to get that far...

So I'm assuming that they haven't suddenly found a way to go at speeds of 1Meg or more using the same wiring?

I'm using a cable connection which is coax so I was just assuming the telephone connection is using the same, if not someone let's hear how??

John.

It occurs to me John, that you might not be as aware of DSL as we are in the states. I think your infrastructure is better.

Copied from Wikipedia:

Typically, the download speed of DSL ranges from 640 kilobits per second (kbit/s) to 3,000, or exceptionally from 128 to 24,000 kbit/s depending on DSL technology and service level implemented. Typically, upload speed is lower than download speed for Asymmetric Digital Subscriber Line (ADSL) and equal to download speed for the rarer Symmetric Digital Subscriber Line (SDSL).

All this is on twisted pair copper originally designed for no more than about 3000 hZ analog audio.

The DSL link is here. The ADSL link above gives a little more detail. But in short they implement a multiplicity of channels, each only 4khz wide. Of course if you are far from the telco central office, the whole thing doesn't work well. But I've always been pretty close over the last 6 years or so -- so it works just fine for me.

Separating the voice from data, at my end, is just a simple passive low pass filter. Every day, I remove that filter for about an hour, and practice "listening" to the data while watching my computer screen. My hope is that I will be able to train myself to interpret the data, eliminating the need for my computer -- the data will simply go straight into my head. I am planning to join up with the circus as "Mr. Computer Head".

There is upper band limit filter that sets the limit for your communication link. This is dynamic and software programmable and is part of your modem. New types of broadband modems have programmable features in EEPROM or Flash. You need not change anything physically.

Service provider thus limits this on the bases of what he has and what you pay for. My high speed line cost double to low speed line. Hence I can opt as per my need and payment capacity.

Ahhh thanks for that link Ken...

Being an analogue designer I normally turn off when the subject talks about digital signal processing etc...

So I was still thinking that the broadband ADSL still used the analogue audio part of the telephone line spectrum - Clearly from that link that's not the case and explains why I couldn't understand how they could use twisted pair down a phone line for such high speed connections...

John

Then there is this DSL filter which is a passband for data and voice as signal separator. Without this the voice will sometime hear a burst of noise if you have a bat ears. I can hear that noise and I am not a batman at all. I have two such lines in my house and one from Government very noisy and one from private subscriber very clean.

Ken

Thanks for the link--enough there to keep me reading for awhile.

Lonnie