# Communication channels - TCP-IP

This is a discussion on Communication channels - TCP-IP ; hello The following questions are not related to Multiplexing, where single channel combines multiple signals into one signal ( and at the other end the original channels are extracted by the process called demultiplexing ). For the sake of argument, ...

1. ## Communication channels

hello

The following questions are not related to Multiplexing, where single
channel combines multiple signals into one signal ( and at the other
end the original channels are extracted by the process called
demultiplexing ). For the sake of argument, let's pretend that there
is no such thing as multiplexing

1)
When talking about broadband medium the word channel is often used. A
single Broadband medium can have more than one channel, where each
channel transmits signals with certain frequencies. This way one
broadband medium can transmit multiple signals at the same time.

* Now does the word channel apply to individual wire in a broadband
medium and thus each channel is actually a wire, capable of
transmitting signals with certain frequency? If that is the case then
for broadband cable to have 20 channels, it must have 20 wires?

* Or are channels only a logical construct, meaning that word channel
only refers to a certain frequency? Thus a single wire can at the same
time transmit more than one signal by having each signal transmitted
at different frequency? That's a bit hard to believe.

2)
* To my understanding, if each channel has its own wire, then
broadband cable with only one wire would not be able to send signals
in both directions at the same time ( I think we call this simplex
transmition )?!

* But if word channel only describes different frequencies, then a
single wire can transmit signals in both directions at the same time.
I find that hard to believe, since ( as little as I know about
physics ) I know that a wire can only have current flowing in one
direction at the time ( as such voltage can also have only one
direction at the time )

3)
I'm not saying it would be practical, but if each channel has its own
wire, then why can't all channels have the same frequency? I know
cable TV operates by sending each program with different frequency and
we can tune TV to certain frequency and watch a particular program
( that was send with that frequency ). Since each channel has its own
wire, then each cable program would be send through different wire and
we could tune into specific program simply by connecting TV receiver
( or something ) with that certain wire ( this way all channels could
send signals with same frequency )?!

thank you

2. ## Re: Communication channels

wrote:

> hello
>
> The following questions are not related to Multiplexing, where single
> channel combines multiple signals into one signal ( and at the other
> end the original channels are extracted by the process called
> demultiplexing ). For the sake of argument, let's pretend that there
> is no such thing as multiplexing
>
> 1)
> When talking about broadband medium the word channel is often used. A
> single Broadband medium can have more than one channel, where each
> channel transmits signals with certain frequencies. This way one
> broadband medium can transmit multiple signals at the same time.
>
> * Now does the word channel apply to individual wire in a broadband
> medium and thus each channel is actually a wire, capable of
> transmitting signals with certain frequency? If that is the case then
> for broadband cable to have 20 channels, it must have 20 wires?
>
> * Or are channels only a logical construct, meaning that word channel
> only refers to a certain frequency? Thus a single wire can at the same
> time transmit more than one signal by having each signal transmitted
> at different frequency? That's a bit hard to believe.

Well, you can easily picture multiple frequency channels operating over
the air, right? Radio channels, TV channels. They can be half duplex
(share the same frequency, but take turns transmitting and receiving )
or full duplex (transmit and receive simultaneously, using seperate
frequencies). So, why not do the same thing over a cable?

Sometimes "channel" refers to a single media stream, and of course media
streams can be separated within any medium, using packet headers to
differentiate the streams.

> 2)
> * To my understanding, if each channel has its own wire, then
> broadband cable with only one wire would not be able to send signals
> in both directions at the same time ( I think we call this simplex
> transmition )?!
>
> * But if word channel only describes different frequencies, then a
> single wire can transmit signals in both directions at the same time.
> I find that hard to believe, since ( as little as I know about
> physics ) I know that a wire can only have current flowing in one
> direction at the time ( as such voltage can also have only one
> direction at the time )

If you consider the use of passband filters, then you can conceive of a
cable being able to send data in both directions over the same coax or
twisted pair, at the same time. In different frequency bands, energy can
flow in either direction. For example, Transmitter A will see a very
high impedance on all frequencies except Frequency 1. So Transmitter A
will only be able to send energy out on the Frequency 1 band. At the
same time, Receiver A connected to that same cable will be blocked from
receiving any energy from the cable, except on the Frequency 2 passband.
So the same cable will be used to send energy in both directions at the
same time, and Transmitter A will not damage Receiver A by blasting it
with too much energy.

Should not be hard to believe. You can certainly do this with cell
phones over the air, right?

Alternatively, the system can share the same band, switching quickly
between transmit and receive, so that in use, a human won't notice. This
describes the original coax Ethernet, for example. Not so easy to do
with voice, though.

> 3)
> I'm not saying it would be practical, but if each channel has its own
> wire, then why can't all channels have the same frequency? I know
> cable TV operates by sending each program with different frequency and
> we can tune TV to certain frequency and watch a particular program
> ( that was send with that frequency ). Since each channel has its own
> wire, then each cable program would be send through different wire and
> we could tune into specific program simply by connecting TV receiver
> ( or something ) with that certain wire ( this way all channels could
> send signals with same frequency )?!

If each channel has a dedicated twisted pair, or dedicated wire and
common ground, in principle you would not need to send the signal over
different RF carriers. For example, twisted pair Ethernet, like
10BASE-T, has separate, dedicated transmit and receive twisted pair
cables between each host and each hub or switch. The signals are
transmitted at baseband. Which means, they use the same carrier
frequency, if you will.

Bert

3. ## Re: Communication channels

wrote:

> hello
>
> The following questions are not related to Multiplexing, where single
> channel combines multiple signals into one signal ( and at the other
> end the original channels are extracted by the process called
> demultiplexing ). For the sake of argument, let's pretend that there
> is no such thing as multiplexing

Oh, and by the way, what I decribed in the previous reply *is*
multiplexing. It's called "frequency division multiplexing," or FDM, as
opposed to "time division multiplexing," or TDM, but multiplexing
nevertheless.

And what is commonly referred to as "wave division multiplexing," WDM,
is the same as FDM exactly. It simply applies to higher frequencies in
the EM spectrum, up in the visible light range, or near there.

All are forms of multiplexing. Different techniques used to share a
single medium among many data streams.

Bert

4. ## Re: Communication channels

On Jun 4, 1:10 pm, kaja_love...@yahoo.com wrote:
> hello

> * Now does the word channel apply to individual wire in a broadband
> medium and thus each channel is actually a wire, capable of
> transmitting signals with certain frequency? If that is the case then
> for broadband cable to have 20 channels, it must have 20 wires?
>
> * Or are channels only a logical construct, meaning that word channel
> only refers to a certain frequency? Thus a single wire can at the same
> time transmit more than one signal by having each signal transmitted
> at different frequency? That's a bit hard to believe.

When thinking about something like cable TV, it is the latter. It is
very possible for a wire to transmit more than one frequency at the
same time. If you take an oscilloscope and look at the signal on the
wire, and and discover that it is not a pure sine wave, but some
warbling thing that does not make any sense (is not periodic), then
technically, that thing is a bunch of sine waves added together that
are changing in time. This was the thesis of the most his eminence,
J.B.J. Fourier, of Fourier Theory http://en.wikipedia.org/wiki/Joseph_Fourier).

> * To my understanding, if each channel has its own wire, then
> broadband cable with only one wire would not be able to send signals
> in both directions at the same time ( I think we call this simplex
> transmition )?!

In fact, it is possible and used every day. This is why, among a many
other situations, a plain old telephone system (POTS) phone has only
two wires instead of four. Both ends of the phone line, the end in
the house as well as the end in the central office, use inductors
(http://en.wikipedia.org/wiki/Inductor) to induce into the wire the
signal that they want on the wire. The signals manifest as charge
distributions along the wire. Normally, if only one side were
inducing, there would be only one charge distribution. But if both
sides induce, then you can think of the sides generating charge
distributions that "fight" with each other (sometimes causing a zero
at some point along the wire, but not other points). The resulting
charge distribution after the "fighting" is nothing more than the the

> * But if word channel only describes different frequencies, then a
> single wire can transmit signals in both directions at the same time.
> I find that hard to believe, since ( as little as I know about
> physics ) I know that a wire can only have current flowing in one
> direction at the time ( as such voltage can also have only one
> direction at the time )

The signal can be, indeed, transmitted in both directions at the same
time. The funny thing about your remark is that you can think of it
at many different levels. Many electrical engineers take the
principle of superposition as obvious. But if you think really hard
about it, you come to realize that electromagnetic waves must be
implemented as distributions of quantized particles of light, each
with there own momentum, so you get into quantum mechanics very
quickly. It is true..if you present a pure DC signal against the wire
at both ends, assuming no noise, there will be a fixed net voltage,
fixed electrical field, and therefore a fix flow in one direction.
When you start jiggling the charges however, you also giggle the flux
of electromagnetic emission, and their movement result in propagation
of (time-varying) fields. What's remarkable is that the moment was
there before, only now, it's time varying, so there is a visible
manifestation as a result at each end.

> I'm not saying it would be practical, but if each channel has its own
> wire, then why can't all channels have the same frequency? I know

The can. Called mulltiplexing.

> cable TV operates by sending each program with different frequency and
> we can tune TV to certain frequency and watch a particular program
> ( that was send with that frequency ). Since each channel has its own
> wire, then each cable program would be send through different wire and
> we could tune into specific program simply by connecting TV receiver
> ( or something ) with that certain wire ( this way all channels could
> send signals with same frequency )?!

Note that there is nothing inherently desirable about using same
frequency in a coaxial cable. Also note that there are modulation
schemes were this is done. amplitude modulation, phase modulation,
quadrature amplitude modulation , etc. http://en.wikipedia.org/wiki/Amplitude_modulation

As for as many wires...I have about 200 channels on my
television...and so, the wire would be about the thickness of my arm.
The cable installer would get very tired lugging such a cable around.

-Le Chaud Lapin-

5. ## Re: Communication channels

wrote in message
> hello
>
> The following questions are not related to Multiplexing, where single
> channel combines multiple signals into one signal ( and at the other
> end the original channels are extracted by the process called
> demultiplexing ). For the sake of argument, let's pretend that there
> is no such thing as multiplexing
>
> 1)
> When talking about broadband medium the word channel is often used. A
> single Broadband medium can have more than one channel, where each
> channel transmits signals with certain frequencies. This way one
> broadband medium can transmit multiple signals at the same time.
>
> * Now does the word channel apply to individual wire in a broadband
> medium and thus each channel is actually a wire, capable of
> transmitting signals with certain frequency? If that is the case then
> for broadband cable to have 20 channels, it must have 20 wires?

No
>
> * Or are channels only a logical construct, meaning that word channel
> only refers to a certain frequency?

Getting warm. This is where frequency division multiplexing comes in

> Thus a single wire can at the same
> time transmit more than one signal by having each signal transmitted
> at different frequency? That's a bit hard to believe.

Not at all hard to believe.

Why would you say that is hard to believe.

>
>
>
> 2)
> * To my understanding, if each channel has its own wire, then
> broadband cable with only one wire would not be able to send signals
> in both directions at the same time ( I think we call this simplex
> transmition )?!
>
> * But if word channel only describes different frequencies, then a
> single wire can transmit signals in both directions at the same time.
> I find that hard to believe, since ( as little as I know about
> physics ) I know that a wire can only have current flowing in one
> direction at the time ( as such voltage can also have only one
> direction at the time )

Volatge is not the issue here.
>
>
> 3)
> I'm not saying it would be practical, but if each channel has its own
> wire, then why can't all channels have the same frequency?

Interference for starters

> I know
> cable TV operates by sending each program with different frequency and
> we can tune TV to certain frequency and watch a particular program
> ( that was send with that frequency ). Since each channel has its own
> wire, then each cable program would be send through different wire and
> we could tune into specific program simply by connecting TV receiver
> ( or something ) with that certain wire ( this way all channels could
> send signals with same frequency )?!

Cable TV as an example. The cable used carries a frequency range, with
different discrete frequency channels within that cable
>
>
> thank you
>

6. ## Re: Communication channels

hiya

>>The following questions are not related to Multiplexing, where single
>>channel combines multiple signals into one signal ( and at the other
>>end the original channels are extracted by the process called
>>demultiplexing ). For the sake of argument, let's pretend that there
>>is no such thing as multiplexing

>
>Oh, and by the way, what I decribed in the previous reply *is*
>multiplexing. It's called "frequency division multiplexing," or FDM, as
>opposed to "time division multiplexing," or TDM, but multiplexing
>nevertheless.

If I understood you guys correctly, you are saying that we can for
example send two signals at the same time through a single wire -
first signal in one direction and second in the opposite direction.
But I don't understand why multiplexing would be required for two
signals to travel in opposite direction. Isn't multiplexing a
technique which combines signals traveling in same direction, into one
signal?
If two signals are traveling in opposite directions, then there is no
need to combine the two signals, since the meeting of two waves
( signals ) along a medium does not alter the shape of individual
waves?!

cheers

7. ## Re: Communication channels

wrote:

> If I understood you guys correctly, you are saying that we can for
> example send two signals at the same time through a single wire -
> first signal in one direction and second in the opposite direction.

Yes. You can even try this in a sink full of water. Fill up a sink. Then
touch the water on the left with one finger, and at the same time, touch
the water on the right with one finger. You will see ripples propagating
at the same time from left to right and right to left. The ripples cross
each other and stay on their previous course. So you can say that the
single medium, in this case a pool of water, is simultaneously carrying
more than one signal.

Electrons can similarly "bump" each other that way, either absorbing or
giving off energy, thereby propagating a signal in both directions.

> But I don't understand why multiplexing would be required for two
> signals to travel in opposite direction. Isn't multiplexing a
> technique which combines signals traveling in same direction, into one
> signal?

No requirement that the signals be in the same direction. "Multiplexing"
is a very broad term. Here is the Wikipedia definition:

"multiplexing (short muxing) is a term used to refer to a process where
multiple analog message signals or digital data streams are combined
into one signal."

I would even more simply say that multiplexing means to share a medium
among many "channels" or "data streams" or whatever type of connection
you care to define. The exact technique may vary, but the essence
remains the same.

> If two signals are traveling in opposite directions, then there is no
> need to combine the two signals, since the meeting of two waves
> ( signals ) along a medium does not alter the shape of individual
> waves?!

But it does alter the energy transferred through the medium. From the
outside looking in, you will have changed the EM patterns, or "signal,"
on that one cable, for example.

Bert

8. ## Re: Communication channels

thank you all for your help

cheers