Page 203 - Integrated Wireless Propagation Models

M a c r o c e l l P r e d i c t i o n M o d e l s - P a r t 2 : P o i n t - t o - P o i n t M o d e l s 181

3.5.1 Ways of Implementation of Models
If we look at the models described in Chaps. 2 and 3 from the ways they are imple­
mented, we can categorize them into four different approaches:

Empirically Based-Young, Okamura, and Hata use the measurement-based graphi­
cal method that predicts the received field strength as a function of various terrain
features and system parameters. Young's measurement data were taken at the micro­
wave frequency range from 150 to 3700 MHz on the East Coast of the United States.37
Okamura's measurement data36 were taken around 150 to 800 MHz in Japan.
Terrain Data-An accurate terrain database for calculating the transmission path
loss over irregular terrain based on the statistical properties of the terrain profile
is needed. The detailed point-to-point version rather than the area version is stud­
ied in this chapter because it provides better prediction accuracy. A portion of the
Longley-Rice procedure is used in conjunction with International Radio Consultative
Committee (CCIR) propagation curves based on a method published by Durkin
in 1977.38 Durkin's computer program generates field strength contours around a
base station by computing path loss estimates along radial lines. Two other pre­
diction techniques, one from the Joint Radio Committee (JRC) of the Nationalized
Power Industries of the United Kingdom and the other from Palmer of the Canadian
Communications Research Centre (CRC)39 are also operating on computer programs
using terrain database. The Lee model is a point-to-point model and has used mixed
theoretical derivations and statistical approaches in predicting the signal strengths
on U.S. Defense Map Agency (DMA) topographic data (1 ox 1 o tape) with a satisfac­
tory prediction as compared with the measured data. 22
Plane Earth-Egli was the first to publish a technique incorporating the plane earth
model, estimating received power by treating the earth as a plane surface and adding
0
a terrain clutter factor to account for surface obstructions. Egli4 estimates the median
path loss in decibels by summing the plane earth reference loss and the median
1
terrain factor. Later, Murphy4 presented a statistical method for predicting trans­
mission loss over the irregular terrain using empirical data from the plains and
mountains of Colorado. A third method was developed by Allsebrook and Parsons
2
in the United Kingdom.4 They designed a prediction technique specifically for urban
environments using data in three British cities. The Allsebrook-Parsons model was
introduced in Chap. 2.
Monograms-Bullington provides a short manual procedure for making estimates
of the average received power using graphs and monograms. His study also contains
a brief summary of vehicular radio propagation.

3.5.2 Features Among Models
Different models have different features. A summary of model features is shown in
Table 3.5.2. There are three major categories: input parameters, propagation factors,
and output treatment parameters. Under each category, is a list of items. We take each
model and check whether covers the items. N means not treated, L means limited
treated, and E means extend treated. The simpler models that can be found in the
table are Carey34 and CCIR 56743 introduced in Chap. 2. Depending on the purpose of
using a particular model, some parameters may not be important. The Lee model has
covered most features and has tried to make a more accurate prediction.