Walsh Codes

Reiterating the utilization of Walsh codes has a direct impact upon the
radio networks ability to carry and transport the various services.With the
introduction of CDMA2000, there are several alterations to the use of
Walsh codes that were previously discussed, but only briefly.
With CDMA2000, the Walsh codes now have variable lengths that range
from 4 to a total of 256, which is an expansion over IS-95 systems that only had 64 codes. The one effect with utilizing variable length Walsh codes is
that if a shorter Walsh code is being used, then it precludes the use of the
longer Walsh codes that are derived from it. For instance, if Walsh code 2 is
used, then it precludes the use of all the Walsh codes in the code tree that
were derived from it.
Table 13-18 helps in establishing the relationship between which Walsh
code length is associated with a particular data rate. However how does this
play into the use of determining the radio network?
For a SR1 and RC1 there are a maximum number of users that have
individual Walsh codes equating to 64, a familiar number from IS-95A.
However, if we were to have a R3 capable base radio with a SR1, phase 1
CDMA2000, and we had a total of 12 RC1 and RC2 mobiles under that sector,
then this would only allow for three data users at 153.6K, or 6 at 76.8
Kbps, 13 at 38.4, 26 at 19.2, or 104 at 9.6 Kbps. Obviously the negotiated
mobile data rate complicates the determination for the total throughput of
traffic levels. The real issue behind this is that the type of data that will be
enabled to be transported over the network has a direct impact on the available
users. If for example, the need were for high-speed data for interactive
video.With a R3 capable mobile, 384K of bandwidth, would not be feasible.

At times it is best to see the previous example in a visual format in order
to better understand the relationship between the short and long Walsh
codes. Table 13-19 shows the relationship, or rather, the Walsh tree, from 4
to 256 Walsh codes and their relative relationship with one another. The
relationship is illustrated in Table 13-20.

Table 13-20 is an illustration of the interaction of a Walsh code and that
of the its higher or lower branches. The Walsh codes that are consumed are
depicted in the shaded area. Also the use of Walsh codes for the various
channels that are associated with CDMA2000 are not included here
because they too draw upon the same Walsh code pool. However, for ease of
illustration, they were left out for the example.
In the example shown in Table 13-20, the use of Walsh code 48, which is
CDMA2000-capable, and is set up for low-speed packet data and voice
applications, precludes the use of high-speed packet data from utilizing this
set of Walsh codes, thereby effectively reducing the sites data handling
capability by 25 percent with the use of a single voice call. Alternatively the use of a single high-speed data session using Walsh code 1 eliminates from
possible use a total of 64/32 Wash codes. Now there is a difference between
both examples; the first is that the data session will end sooner, at least in
concept, then the voice call, thereby replenishing the Walsh code pool.
Another very important issue regarding the Walsh code pool is that with
3X channels, the same Walsh code is used for all three carriers associated
with the 3X platform.
Therefore, based on the expected traffic mix that is anticipated for the
system, the choice of how to deploy the services relative to the carriers is
important. To be more blunt, if there is a 50/50 mix between packet and
voice traffic and the packet usage may be 70 Kbps or higher, then it is advisable
that when deploying CDMA2000, a separate channel is used for packet
data only thereby preserving the imbedded voice platforms and of course
throughput. 528