Comments on: Cell Phone Towers http://www.visiognomy.com/diagrams/archives/2005/02/16/cell-phone-towers/ Life in diagrams. en Tue, 6 Jan 2009 02:13:27 +0000 http://wordpress.org/?v=1.2.2 by: James Wetterau http://www.visiognomy.com/diagrams/archives/2005/02/16/cell-phone-towers/#comments Tue, 22 Feb 2005 10:57:50 -0500 6:15@http://www.visiognomy.com/diagrams Ah but in New York (and most of the US I think) often cellphones are not just a regular radio. Many cellular networks use CDMA, which uses a spread spectrum approach in which each phone uses an entire range of frequencies. Each station (the towers and the cellphones, I guess) is assigned a code, and that code is broadcast across the frequencies. The phones and towers can distinguish that code from among all the codes simultaneously being broad- cast as long as the total power of the other calls does not exceed the power of the station of interest by too much. The approach is to rely on the sum of the data having certain properties that allow each distinct code to be recognized from the total signal by a mathematical operation – the dot product. Once you can recognize the code, you can recognize the call. (I don’t know how this is done – perhaps by its FM characteristics?) This has advantages of other ways of carving things up (e.g. by frequency) because other methods have to be prepared a worst-case scenario and try to calculate the minimum useable amount of bandwidth ahead of time, or else alllocate it and then retrieve it regularly, managing that process dynamically, whereas CDMA bandwidth doesn’t have to be managed. It will achieve a good signal-to-noise ratio under most ordinary conditions. Once codes are assigned, they’re good forever, and new stations can be added at will without worrying about managing the use of the spectrum or time slots. Quality of the signal remains fairly consistent as the number of simultaneous users in a cell changes, unless the system is simply overwhelmed by too many simultaneous users, which in practice seldom happens. Ah but in New York (and most of the US I think) often
cellphones are not just a regular radio. Many cellular
networks use CDMA, which uses a spread spectrum approach
in which each phone uses an entire range of frequencies.

Each station (the towers and the cellphones, I guess) is
assigned a code, and that code is broadcast across the
frequencies. The phones and towers can distinguish that
code from among all the codes simultaneously being broad-
cast as long as the total power of the other calls does
not exceed the power of the station of interest by too
much. The approach is to rely on the sum of the data
having certain properties that allow each distinct
code to be recognized from the total signal by a
mathematical operation – the dot product.

Once you can recognize the code, you can recognize
the call. (I don’t know how this is done – perhaps
by its FM characteristics?)

This has advantages of other ways of carving things up
(e.g. by frequency) because other methods have to be
prepared a worst-case scenario and try to calculate the
minimum useable amount of bandwidth ahead of time, or else
alllocate it and then retrieve it regularly, managing that
process dynamically, whereas CDMA bandwidth doesn’t have to
be managed. It will achieve a good signal-to-noise ratio
under most ordinary conditions. Once codes are assigned,
they’re good forever, and new stations can be added at will
without worrying about managing the use of the spectrum or
time slots. Quality of the signal remains fairly consistent
as the number of simultaneous users in a cell changes,
unless the system is simply overwhelmed by too many
simultaneous users, which in practice seldom happens.

]]> by: James Wetterau http://www.visiognomy.com/diagrams/archives/2005/02/16/cell-phone-towers/#comments Tue, 22 Feb 2005 11:01:57 -0500 7:15@http://www.visiognomy.com/diagrams Oh, I meant to add these three URLs that explain CDMA and also provide some illustration of the process: http://en.wikipedia.org/wiki/CDMA http://www.umtsworld.com/technology/cdmabasics.htm http://www.cdg.org/technology/cdma%5Ftechnology/a%5Fross/cdmarevolution.asp Oh, I meant to add these three URLs that explain CDMA
and also provide some illustration of the process:

http://en.wikipedia.org/wiki/CDMA
http://www.umtsworld.com/technology/cdmabasics.htm
http://www.cdg.org/technology/cdma%5Ftechnology/a%5Fross/cdmarevolution.asp

]]> by: adam http://www.visiognomy.com/diagrams/archives/2005/02/16/cell-phone-towers/#comments Tue, 22 Feb 2005 14:07:34 -0500 8:15@http://www.visiognomy.com/diagrams Your comments are well taken. But it’s still a regular radio, right? It’s a frequency hopping digital radio signal with a digital code attached to each transmission, but the fact that you can’t tune into the call with a standard analog radio receiver doesn’t change the qualities of the underlying transmission medium. Your comments are well taken. But it’s still a regular radio, right? It’s a frequency hopping digital radio signal with a digital code attached to each transmission, but the fact that you can’t tune into the call with a standard analog radio receiver doesn’t change the qualities of the underlying transmission medium.

]]> by: James Wetterau http://www.visiognomy.com/diagrams/archives/2005/02/16/cell-phone-towers/#comments Tue, 22 Feb 2005 16:53:01 -0500 9:15@http://www.visiognomy.com/diagrams It’s not just frequency hopping – it’s broadcasting on multiple frequencies simultaneously. By design, the amount of power it uses on each frequency is low – below the noise threshold for ordinary radio broadcasting – but if you tune in all the frequencies of interest you can pull out this low signal as a weak but consistent variation from the almost random background noise of each frequency. The code makes this consistently achievable (assuming I understand correctly.) I guess the answer would be that it’s a regular radio with two main differences from a walkie-talkie. 1) Instead of a strong signal on one unique frequency, it uses a very weak signal spread across a broad frequency band. 2) Instead of an analog signal, it encodes voice as a digital signal, which it then broadcasts with a unique binary code signature. The point of 1 and 2 is that interference is not an issue as long as the signal is not disastrously overwhelmed. It’s a multi-frequency radio with a special purpose encoding/ decoding processor, customized to have a unique signature for the phone. It’s not just frequency hopping – it’s broadcasting on multiple
frequencies simultaneously. By design, the amount of power it
uses on each frequency is low – below the noise threshold for
ordinary radio broadcasting – but if you tune in all the
frequencies of interest you can pull out this low signal as a
weak but consistent variation from the almost random background
noise of each frequency. The code makes this consistently
achievable (assuming I understand correctly.)

I guess the answer would be that it’s a regular radio with
two main differences from a walkie-talkie.

1) Instead of a strong signal on one unique frequency, it
uses a very weak signal spread across a broad frequency band.
2) Instead of an analog signal, it encodes voice as a digital
signal, which it then broadcasts with a unique binary code
signature.

The point of 1 and 2 is that interference is not an issue
as long as the signal is not disastrously overwhelmed.

It’s a multi-frequency radio with a special purpose encoding/
decoding processor, customized to have a unique signature for
the phone.

]]>