Page 404 - Integrated Wireless Propagation Models

382 C h a p t e r S i x

strength of an interested model, such as digital terrain maps, a few runs of measured
data, and so on, and be sure it meets our requirement. Also understand what the out­
puts are from the prediction models.
Once the prediction model is chosen, we first have to find the path-loss slope and
intercept in the area or in different angles from a site-specific base station from the pre­
diction model. Determine the gains and losses over the radio path before reaching the
mobile receiver. Then the noise floor is calculated, including the thermal noise, equip­
ment noise, human-made noise, and interference, and the C/N or C/I can be found.
Taking the link margin into account, we can determine the (C!N) r To use the planned
p
(CIN) 1 as a parameter, we can plot the boundary around each base station. At this point,
P
we adjust and move around the locations of the base stations to see the coverage of the
system. Those planned base station sites will be rented or negotiated if there is a zoning
issue. Also, we can install the sites owned by the antenna tower companies to be the first
choices. Whenever base stations, are moved, the effectiveness of interference in the sys­
tem changes. The HO areas are critical and need to be well planned to avoid call drops.
Also, the HO regions should be as small as possible so that capacity will not be hurt.
M
6.6.2.2 For i crocell Systems
In the microcell system, we have to realize that the building structures in microcell sys­
tems, the unique landscape of the small area beging to be a key component in the model.
The building layout along the streets and in each street block can be used to find the
building density in each square block around four streets. The prediction model can
find the path loss of the radio wave due to the building blocks. The small-scale terrain
tour map may be used to calculate the effective antenna gain or shadow loss. Sometime
if we do not have enough information, we may use the macrocell model to predict the
path loss. The deviation between the predicted results and the measured data may be
still within 8 dB in 60 percent of CDF, which is acceptable.

(
6.6.2.3 For In-Building P icocell)
The Lee in-building model is a simple and cost-effective prediction model. The layout
of the inside as well as the outside of the building is required. The wall material is must
be known. The attenuation factor of the walls comes from the wall material. A few runs
of measured data for the same floor, inter-floor, inter-building, and outside building
should be collected. All the path-loss slopes are come from the measured data. If no
measured data are available, then the default values can be used. Because the measured
data are used for enhancing the model, the deviation between the predicted results and
measured data should be less than 6 to 8 dB in 60 percent of CD F. The ray-tracing model
and the FDTD model are physically based prediction models. They can be used when
no measured data are available. However the input data for running these models are
more complicated. Some models need 30 display. The costs and time needed are high
and making the models is not very attractive. These models are good to use to find the
physical explanations of many ray path phenomena. There are some military applica­
tions of using these physically based prediction models.

6.6.2.4 For the Lee Comprehensive Model
The Lee comprehensive model can provide an overall signal coverage map of a wireless
system in a large operating area. Because this comprehensive model integrates three
cell size-specific Lee models, this coverage map can include the macrocells, microcells,
and in-building (picocell) cells. This comprehensive model is very useful for planning