Page 285 - Integrated Wireless Propagation Models

I n - B u i l d i n g ( P i c o c e l l ) P r e d i c t i o n M o d e l s 263

Many in-building propagation research material and predictions have never been pub­
lished before and are disclosed in this chapter. The first feature that impacts communi­
cation is assuming a clear space loss between the transmitter and receiver. The second
feature is the loss due to reflections from interior and exterior walls. The third feature is
using different propagation loss formulas for regular rooms (such as conference rooms
or offices) and special rooms (such as elevators and utility rooms). This model provides
an improved prediction by allowing one to understand the effect of the building envi­
ronment on the propagation characteristics of a mobile receiver. The prediction is done
basically along a radial by applying different formulas while passing through different
environments. Because there are many different propagation paths from the transmit­
ting antenna to the receiving antenna, the paths exhibit many different starting direc­
tions at the transmitting antenna and many different directions of arrival at the receiving
antenna. At any point, the signal can be higher or lower than the free space loss due to
the multipath. However, the average signal loss is always higher than free space. The
measured data were collected by moving the transmitter or the receiver over a spatial
area (often running in a circular path) and then averaged by following the traditional
1
1
method to find the local mean, as stated in Sec. . 6.3. . The area had linear dimensions
of 10 to 20 wavelengths for obtaining the local mean signals. These local means can
1 1
remove the rapid variation of the received signal. 7• 8
Once the room penetration loss characteristics are derived, the optimized place­
ment of antenna can be determined by moving the mobile around at each of different
antenna locations. In general, the mobile communication channel often consists of a few
strong paths combined with a number of weaker paths; the indoor channel exhibits
1
similar characteristics. 9 However, the prediction from the model is based only on the
average of the signal strength received by the mobile receiver.

5.2 The Lee I n - B u i l d i n g Prediction Model
The potential implementation of an in-building prediction model for the wireless local
area networks (LAN) and personal communication services (PCS) inside buildings
requires a thorough understanding of signal propagation in a building. In this section, the
Lee in-building (Picocell) prediction model is presented. This prediction model focuses
on a single floor of a building but is also applicable to different floors of the building. This
model is also applicable to the through-building propagation loss by applying the same
principle. The validation of this model was done in two different buildings of similar
construction in the 900-MHz band. A special feature of this model is its ability to handle
different types of obstructions. The model is validated by gathering the measured data for
a specific floor of a building and comparing them with the predicted values. The standard
deviation between measured data and predicted values is within 5 dB.

2
5. . 1 Derivation of Close-I n Distance for the In-Building Model
In this chapter, the near-in distance is derived for the microcell model. Here, we are
dealing with, a different signal propagation environment. It is an in-building environ­
ment, so it is a close-in environment. Since the microcell environment is an open envi­
ronment, the near-in distance used in the microcell environment can not be applied to
this close-in environment. Therefore, we have derived another propagation distance
that is from the base station to a short distance at which the signal still maintains strong
can be treated as a free space path-loss signal. We name the distance the close-in distance.