Code Division Multiple Access (CDMA)

With CDMA, neither
the time domain nor the frequency domain are subdivided. Rather, all users
share the same radio frequency at the same time. This approach obviously
means that all users interfere with each other. Such interference would be
intolerable if the radio frequency bandwidth were limited to just the bandwidth
that would be needed to support a single user. To overcome this difficulty,
CDMA systems use a technique called spread spectrum, which
involves spreading the signal over a wide bandwidth. Each user is allocated
a code or sequence and the bit rate of the sequence is much greater than
the bit rate of the information being transmitted by the user. The information
signal from the user is modulated with the sequence assigned to the
user and, at the far end, the receiver looks for the sequence in question.
Having isolated the sequence from all of the other signals (which appear
as noise), the original user’s signal can be extracted.
TDMA systems have a very well-defined capacity limit. A set number
of channels and a set number of time slots exist per channel. Once all
time slots are occupied, the system has reached capacity. CDMA is somewhat
different. With CDMA, the capacity is limited by the amount of
noise in the system. As each additional user is added, the total interference
increases and it becomes harder and harder to extract a given user’s
unique sequence from the sequences of all the other users. Eventually,
the noise floor reaches a level where the inclusion of additional users
would significantly impede the system’s capability to filter out the transmission
of each user. At this point, the system has reached capacity.
Although it is possible to mathematically model this capacity limit, exact
modeling can prove a little difficult, since the noise in the system
depends on factors such as the transmission power of each individual
mobile, thermal noise, and the use of discontinuous transmission (only
transmitting when something is being said). By making certain reasonable
assumptions in the design phase, however, it is possible to design a CDMA system that provides relatively high capacity without significant
quality degradation.
IS-95/J-STD-008 is the only widely deployed CDMA system for mobile
communications. This system uses a channel bandwidth of 1.23 MHz and is
an FDD system. The fact that the bandwidth is 1.23 MHz means that the
total system bandwidth (typically, 10 MHz, 20 MHz, or 30 MHz) can accommodate
several CDMA radio frequency (RF) channels. Therefore, like
TDMA, IS-95 CDMA also uses FDMA to some degree. In other words,
within a given cell, more than one RF channel may be available to system
users.
A significant advantage of CDMA is the fact that it practically eliminates
frequency planning. Other systems are very sensitive to interference, meaning
that a given frequency can be reused only in another cell that is
sufficiently far away to avoid interference. In a commercial mobile communications
network, cells are constantly being added, or capacity is being
added to existing cells, and each such change must be done without causing
undue interference. If interference is likely to be introduced, then retuning
of part of the network is required. Such retuning is needed frequently and
can be an expensive effort. CDMA, however, is designed to deal with interference
and, in fact, it allows a given RF carrier to be reused in every cell.
Therefore, there is no need to worry about retuning the network when a
new cell is added.