Page 152 - Integrated Wireless Propagation Models

130 C h a p t e r T h r e e

August October December
Time
FIGURE 3.1.5.2.10 The propagation characteristics based on seasons.

As we have mentioned, the fourth dimension, time, is another important morphol­
ogy. For example, leaves on trees change sizes based on season. Some forests will affect
the signal from different propagation characteristics as time passes. Sometimes the
changes from these effects can be swung up to double digits in decibels. This innovation
includes a feedback loop to incorporate the changes of time. It is implemented in two
different steps. First, a sensor is placed to monitor the system's behavior and to provide
the feedback loop. Once the sensor triggers the loop, the system characteristics will be
tuned to the new status based on the criteria. For example, in dense foliage areas in
Atlanta, the propagation loss can swing 10 dB from summer to a cold winter. The sea­
sonal change spans about six months or more. The sensor placed in the field can provide
feedback on the measured data every month or every other month. The prediction model
can be tuned according to the changes. Once the part of propagation path loss is tuned,
as shown in Fig. 3.1.5.2.10, the associated frequency plan, coverage plan, and capacity
plan for the designing system can be also tuned accordingly. Therefore, the system will
always have a parameter of time as one of the inputs and adjust it dynamically.
The slope of the path loss occurring each month (or at any time frame) is different
since the morphology might be sensitive to weather. For example, leaves are in full
bloom in the summertime and gradually fall to become thinner layers. When winter
comes, there are no tree leaves, and the coverage area will be different, except for pine
trees the leaves, of which, called pine needles, will never fall in the wintertime. The
system has a program to deal with this situation.
First based on the peculiarity of the area under test to update the propagation model,
then the coverage, the best servers (base stations), the interference issues, the handoffs
and the system capacity together with dimensioning other network elements (BSC,
MSC, and so on) will be predicted from the model.
In the summertime, as we have mentioned, the whole forest can be densely covered
with leaves and the impact on propagation can be significant. Especially when the dimen­
sions of the pine tree needles are very close to the quarter, half, or one wavelength of radio
wave, major signal loss will be encountered due to the absorption of the propagation