durex是什么无线传输模型及公式_英文版

2021年1月21日发(作者：mists)
Appendix B: NIR Propagation Models

Introduction
Accurate prediction of the electromagnetic radiation power is a key to accurate and cost
effective wireless network planning. However, an exact analytic evaluation of the field
strength is in general a highly computationally intensive process, and even worse,
requires knowledge of the environment to an accuracy of a fraction of the wavelength.
That is the reason that in practice various models for signal strength prediction are used.
Each model is optimized for a specific environment, specific frequency bands, and also for
the available data.
Within the NIR, a selection of the most commonly used models was implemented. The
operator can choose the models according to the specific problem in hand. The NIR offers
the flexibility to define a model per customer or customer group, thus adapt the model to
the problem and the environment.
The NIR also implements a set of models for fading, thus assisting the planner in defining
the margins needed to reach his/her coverage and availability goals. In the current version
of NIR, a new family of propagation models is introduced. These models caters for
principal component of three dimensional diffraction and reflection, thus providing
accuracy close to full ray tracing with a fraction of the computation load.
This appendix is arranged as follows: the propagation models supported by the NIR are
described in the first part. The second part describes the models used for fade margin
calculation, the third part describes some guidelines how to choose the right model.
Propagation Models
DTM Independent Models
The simplest set of models is the DTM (Digital Terrain Map) independent family of models.
Those models do not require any knowledge of the specific environment details, but rather
use general attributes of the environment. Those models are generally used in case a
DTM is not available or is not needed (e.g. studies, estimates for bids, primary planning
etc.)
Free Space
The free space model is applicable in case there are no obstacles, reflections and other
media irregularities between the transmitter and receiver. The model takes into account
only the “wave divergence”, that is the fact the electromagnetic radiation propagates in

spherical waves emanating from the source, and its power flux density is reduced as a
function of the distance squared. The path loss
L
FS

is given by:

where
λ
is the wavelength, and R the range.
When the path loss is expressed in dB, the frequency in GHz and the distance in meters,
the free space loss can be written as:

Dual Slope
The dual slope model assumes a flat earth between transmitter and receiver, as seen in
Figure B.1.

Figure B.1 Dual Slope
The model looks at the signal strength resulting from the coherent addition of the direct
signal and the signal reflected from the ground. As a matter of fact, this signal fluctuates,
as a function of range very heavily as the signals are added constructively or destructively.
However, this model takes only the envelope of the sum signal, resulting in the following
expression for the path loss
L
DS
:

where:

with
ht
and
hr
are the transmitter and receiver antenna highest respectively, and
λ
is the
wavelength.
HATA Models
In fact this is a family of models is based on the work of Okumura and Hata. They are all
empirical models based on measurements in various environments, 900MHz and
1800MHz frequencies and different antenna heights. As the results are not based on
theoretical models, their applicability is limited in frequency band (150 MHz
–
1500 MHz),
antenna heights (up to 200m) and link distance (less than 20km). The models do not take
into account the specific features of the terrain; rather there are different models for
various environments. In all of the models given below the frequency
f
is expressed in
MHz, the range,
R,
is in km, and the antenna heights,
hr
and
ht
are in meters.
Hata Open
This model is for open areas, villages etc. In this model, the path loss is given by:

Hata Suburban
This model is mostly suitable for suburban areas, small buildings and some vegetation.
In this model the path loss is given by:

With
a
(
hr
) is as given in (4).
Hata Urban
For urban areas, the path loss is given by:

Cost Hata
This is an adaptation of the Hata Urban models, made by the European Cost-231
committee, to adapt the model for the 1500-2000 MHz frequency range:

with
Cm
given by:

IEEE MMDS Models
Also known as the SUI (Stanford University Interim) models, those models are similar to
the Hata models, being based on empirical measurements, performed by AT&T Wireless
in 95 cities around the USA. The measurements were made in 1.9GHz. The models were
adopted by the IEEE 802.16 working group during the development of the IEEE 802.16
standard development.
Like the Hata models, the IEEE MMDS is a family of 3 models for different terrains.
The maximum path loss category is hilly terrain with moderate-to-heavy tree densities
(Category A). The minimum path loss category is mostly flat terrain with light tree
densities (Category C). Intermediate path loss condition is captured in Category B. The
average path loss is given by:

where
A
= 20 log10(4
π
d
0
/
λ
) (
λ
being the wavelength in m),
γ
is the path-loss exponent
with
γ =
(a
–
b h
b
+ c / h
b
) for h
b
between 10 m and 80 m (h
b
is the height of the base
station in m), d
0
= 100m and a, b, c are constants dependent on the terrain category.

The above path loss models is based on published literature for frequencies close to 2
GHz and for receive antenna heights close to 2 m. In order to use the model for other
frequencies and for receive antenna heights between 2 m and 10 m, correction terms
have to be included. The path loss model (in dB) with the correction terms would be

where L
MMDS
is the path loss given in (8),
Δ
P
Lf
(in dB) is the frequency correction term
given by:

with
f
, the frequency in MHz, and
Δ
P
Lh
(in dB) is the receive antenna height correction
term given by

DTM Dependent Models
While all the models mentioned previously do not take into account specific features of the
environment and do not make use of DTM, a more accurate (per point) prediction can be
achieved with models that do use the information about the specific environment included
within the DTM. The DTM is essential for final planning.
Single Knife-Edge
The single knife-edge model deals with the diffraction caused by a single obstacle, as
depicted in Figure B.2.

Figure B.2 Single Knife-Edge
The total path loss is calculated according to the free space loss, as in (2), adding to that
(in dB) a diffraction factor,
F
, calculated by the Fresnel Integral:

In fact the NIR calculates an approximation to that integral given by:

v
is the clearance parameter, which signifies what part of the Fresnel zone, depicted in
Figure 2, is obstructed:

where
H
is the height of the obstacle above the line connecting the transmitter and
receiver,
λ
is the wavelength,
d
1 and
d
2 are the distances of the transmitter and receiver
from the obstacle.
JRC Model
This model, developed by the UK Joint Radio Committee, is a modification of the Knife-
Edge model and it takes into account reflections from the ground. It is a combination of the
Knife-Edge model and the dual slope model described above. More specifically, the path
loss is given by:

where
L
FS

and
L
DS

are the path loss predicted by the free space and dual slope models
respectively, as in equations (2) and (3).
Multiple Knife Edge Models
Calculation of multiple diffraction propagation, resulting from several obstacles, is
computationally difficult. Consequently, several approximation methods exist in literature.
Two of them were implemented in the NIR:
Bullington Method
In this approximation two highest (lowest clearance) obstacles are taken, and replaced by
an equivalent obstacle, as illustrated in figure B.3.

Figure B.3 Multiple Knife-Edge, Bullington
The path loss is calculated with the new obstacle as per Eq. (10).
Deygout Method
This approximation takes three obstacles at the most. Starting from the highest obstacle
the method calculates the additional diffraction loss,
F1
, resulting from the highest
obstacle. Then, the procedure is repeated again (in certain conditions) for the paths
between the transmitter and the principle obstacle and from the principle obstacle to the
receiver, as shown in Figure B.4.
Rounded Obstacle Model
Diffraction from a rounded obstacle is more severe than that from a knife-edge type of
obstruction. The NIR offers two models for this case, the empirical model and the ITU
model, in both a check-up is made to confirm if an obstacle detected by the software is a
knife edge or, if the situation is better described by a rounded obstacle case.

Figure B.4 Multiple Knife-Edge, Deygout
Empirical RO
In this model the NIR uses the rounded obstacle curve, as described in ITU-R rec. P.530.
The curve is relevant for smooth earth situation.
Rounded obstacle
–
ITU
This model implements the curve recommended in ITU-R rec. P.530: