Bandwidth increasing with data rate

Can someone explain to me why a higher data-rate requires a higher
bandwidth as opposed to higher *frequency* of transmission.

Thanks,
--Shafik

Shafik wrote:

>Can someone explain to me why a higher data-rate requires a higher
>bandwidth as opposed to higher *frequency* of transmission.
>
>Thanks,
>--Shafik
>  
>
Let's say you are communicating down a fibre, with a laser. A really 
good, single mode laser. Now, you set a bandwidth constraint somewhere 
near zero. You have lots of frequency, and very little bandwidth. How 
much data can you transfer? Not much. Modulate the amplitude to say 
something, and you break your bandwidth constraint. Change the 
frequency, and you break your bandwidth constraint. Pretty much anything 
else you could do to modulate the laser is doing something with the 
light's amplitude and its phase or frequency. So much frequency, so 
little data transfer.

Now reverse the above. Modulate a low frequency radio signal, with a 
reasonably wide bandwidth constraint. Now you can fiddle around with the 
signal in complex, information bearing, ways without breaking your 
bandwidth constraint.

Look up Shannon's channel capacity work for a more rigorous analysis. :-)

Regards,
Steve

Steve Underwood <steveu@dis.org> writes:

> Look up Shannon's channel capacity work for a more rigorous analysis. :-)

  R = W * log(1 + P/N)

If bandwidth stays constant, we can increase the capacity (i.e., data
rate) by increasing the SNR. Thus the OP's assertion is actually
false, i.e., bandwidth does not necessarily have to increase in order
to increase the data rate.
-- 
%  Randy Yates                  % "Watching all the days go by...    
%% Fuquay-Varina, NC            %  Who are you and who am I?"
%%% 919-577-9882                % 'Mission (A World Record)', 
%%%% <yates@ieee.org>           % *A New World Record*, ELO
http://home.earthlink.net/~yatescr

Randy Yates wrote:

>Steve Underwood <steveu@dis.org> writes:
>
>  
>
>>Look up Shannon's channel capacity work for a more rigorous analysis. :-)
>>    
>>
>
>  R = W * log(1 + P/N)
>
>If bandwidth stays constant, we can increase the capacity (i.e., data
>rate) by increasing the SNR. Thus the OP's assertion is actually
>false, i.e., bandwidth does not necessarily have to increase in order
>to increase the data rate.
>  
>
Of course, what he said was not strictly true. A narrow band with 
massive SNR has a high capacity. However, I think his confusion was why 
increasing the carrier frequency brings no direct capacity benefit at 
all. Of course, it brings various real world benefits, which Shannon 
does not address - in the real world we manage to get rather more down 
an optical fibre than through any microwave link. :-)

Regards,
Steve

Steve Underwood wrote:

(snip)

> Of course, what he said was not strictly true. A narrow band with 
> massive SNR has a high capacity. However, I think his confusion was why 
> increasing the carrier frequency brings no direct capacity benefit at 
> all. Of course, it brings various real world benefits, which Shannon 
> does not address - in the real world we manage to get rather more down 
> an optical fibre than through any microwave link. :-)

Also, in many cases the SNR is better at higher frequencies.
Analog cell phones in the 800MHz region can transmit with
a power level less than one watt without much trouble.
For frequencies below 1MHz the background electrical noise
would be too high for 1 watt to work.

Also, there is relatively more bandwidth available at higher
frequencies.   FM radio is in the 100MHz region as there aren't
many 200kHz wide channels near the AM 1MHz region.

-- glen

On Tue, 31 Aug 2004 14:30:48 +0800, Steve Underwood <steveu@dis.org>
wrote:

>Of course, what he said was not strictly true. A narrow band with 
>massive SNR has a high capacity. However, I think his confusion was why 
>increasing the carrier frequency brings no direct capacity benefit at 
>all. Of course, it brings various real world benefits, which Shannon 
>does not address - in the real world we manage to get rather more down 
>an optical fibre than through any microwave link. :-)
>
>Regards,
>Steve

That's because the signals that are sent down that optical fiber use
substantially more bandwidth than most.   The high carrier frequency
is related to the technology that allows one to do that, but I don't
think the carrier frequency really has anything to do with the
capacity.   Naturally it is a bit difficult to have a 10MHz wide
signal with a 1MHz carrier frequency, but you know what I mean...  ;)


Eric Jacobsen
Minister of Algorithms, Intel Corp.
My opinions may not be Intel's opinions.
http://www.ericjacobsen.org

On Wed, 08 Sep 2004 08:11:29 GMT, glen herrmannsfeldt
<gah@ugcs.caltech.edu> wrote:

>Also, in many cases the SNR is better at higher frequencies.
>Analog cell phones in the 800MHz region can transmit with
>a power level less than one watt without much trouble.
>For frequencies below 1MHz the background electrical noise
>would be too high for 1 watt to work.
>
>Also, there is relatively more bandwidth available at higher
>frequencies.   FM radio is in the 100MHz region as there aren't
>many 200kHz wide channels near the AM 1MHz region.
>
>-- glen

Yikes!   "Higher" is definitely in the eye of the beholder.   Nowadays
I'm used to talking about 2.5GHz and 5-6GHz operation, and even 30GHz
and 60GHz!   800MHz isn't high at all by that standard.

I'd hope that any cell phone would be transmitting with a lot less
than a Watt, especially if it has an antenna that you hold next to
your head.   I think generally they try to keep those guys to 250mW or
so.  There's a Specific Absorption Rate limitation for soft tissue for
safety issues and I think it works out to limiting things that radiate
near your head to 250mW or 500mW or so, don't remember the exact
details but some of our friends in that industry may recall more
perfectly than I.


Eric Jacobsen
Minister of Algorithms, Intel Corp.
My opinions may not be Intel's opinions.
http://www.ericjacobsen.org

Eric Jacobsen wrote:

>On Tue, 31 Aug 2004 14:30:48 +0800, Steve Underwood <steveu@dis.org>
>wrote:
>
>  
>
>>Of course, what he said was not strictly true. A narrow band with 
>>massive SNR has a high capacity. However, I think his confusion was why 
>>increasing the carrier frequency brings no direct capacity benefit at 
>>all. Of course, it brings various real world benefits, which Shannon 
>>does not address - in the real world we manage to get rather more down 
>>an optical fibre than through any microwave link. :-)
>>
>>Regards,
>>Steve
>>    
>>
>
>That's because the signals that are sent down that optical fiber use
>substantially more bandwidth than most.   The high carrier frequency
>is related to the technology that allows one to do that, but I don't
>think the carrier frequency really has anything to do with the
>capacity.   Naturally it is a bit difficult to have a 10MHz wide
>signal with a 1MHz carrier frequency, but you know what I mean...  ;)
>  
>
Of course.

Few practical situations allow the bandwidth to exceed a few percent of 
the carrier frequency. Phone line modems are probably the main 
exception, often using bandwidths exceeding the carrier frequency. So, a 
prerequisite for being able to use a certain bandwidth is generally to 
have a carrier frequency many times that bandwidth. I think that is 
nearly obvious, which tends to give the layman the intuitive feel than 
carrier frequency should somehow be fundamentally related to information 
capacity, which it isn't. I think most of us found it surprising when we 
first learned that carrier frequency was irrelevant.

What's wrong with a signal extending from -4MHz to +6MHz, anyway? 
Complex sampled at 12MHz it works out just fine. :-)

Regards,
Steve

Steve Underwood wrote:

   ...

> What's wrong with a signal extending from -4MHz to +6MHz, anyway? 
> Complex sampled at 12MHz it works out just fine. :-)

Yeah, but I haven't yet figured out what kind of wire to use to send
that signal to the sampler.

Jerry
-- 
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������

"Eric Jacobsen" <eric.jacobsen@ieee.org> wrote in message
news:4140ec47.259424359@news.west.cox.net...
>
> Yikes!   "Higher" is definitely in the eye of the beholder.   Nowadays
> I'm used to talking about 2.5GHz and 5-6GHz operation, and even 30GHz
> and 60GHz!   800MHz isn't high at all by that standard.
>
> I'd hope that any cell phone would be transmitting with a lot less
> than a Watt, especially if it has an antenna that you hold next to
> your head.   I think generally they try to keep those guys to 250mW or
> so.  There's a Specific Absorption Rate limitation for soft tissue for
> safety issues and I think it works out to limiting things that radiate
> near your head to 250mW or 500mW or so, don't remember the exact
> details but some of our friends in that industry may recall more
> perfectly than I.
>

Hello Eric,

Standard CDMA phones operate at 200mW or less. Other phones worked up to
600mW.  The FCC's MPE (Maximum Permissible Exposure) spec which gives the
maximum exposure limits works out to be in the neighborhood of a few mW per
cm^2 (it is frequency band dependent). Of course this spec is based on long
exposures, so one should keep their calls short in duration if they are
worried. Basically in the 300-1500MHz band the 6 minute exposure is f/300
mW/cm^2. So a 900MHz phone allows for 3mW/cm^2.  For the same band the
30minute exposure is f/1500 or 1/5 that of the 6 minute exposure.  For PCS
phones (specifically the band ranging from 1.5GHz up to 100GHz) the MPE is
5mW/cm^2 and 1 mW/cm^2 for 6 and 30 minute exposures respectively.  The
exposure spec is based upon tissue heating and nothing more sophisticated.
Statistical studies of brain tumers have failed to show anything conclusive,
so as they say, the jury is still out about cellphones and carcinogenisis.
However we can cap the carcinogenisis rate with the otherwise normal
occurance rate of such tumers. WE say this because the cellphone experiment
has been carried out with a large population over a long period of time with
no statistically significant increase in observed carcinogenisis.  We
certainly know in the early days of phones with 5W of output that they
caused cataracts when the phone was held in front of the face and the
antenna is right in front of the eyes. Here the 5W cooks the corneas just
like heating an egg turns the "egg white" from clear to white.

Clay S. Turner