通信人家园

标题: 移动无线系统传播环境建模的现代处理  [查看完整版帖子] [打印本页]

---

时间:  2004-12-18 13:04

作者: mick     标题: 移动无线系统传播环境建模的现代处理

IEEE COMSOC. SURVEY vol.3, no.3
Modern Approaches in Modeling of Mobile Radio Systems Propagation Environmen
t
Aleksandar Neskovic, Natasa Neskovic, and George Paunovic,  University of Be
lgrade
移动无线系统传播环境建模的现代处理
Abstract
In this article a review of popular propagation models for wireless communic
ation channels is given. Macrocell, microcell, and indoor prediction methods
are considered separately. Advantages and disadvantages of these models are
discussed. Also, some practical improvements of the existing models as well
as some new models are given.
摘要：
本文对流行的无线通信信道模型进行了综述。对宏蜂窝、微蜂窝以及室内预测模型分别
进行了讨论。讨论了这些模型的优缺点。此外，给出了对这些已有模型的一些实际改进
和一些新的模型。
Over the few past decades, radio communication systems underwent extensive d
evelopment. The demands that a radio system must fulfill are greater by the
day. Having in mind good quality and cost effective solutions, a new radio s
ystem must be designed carefully from the very beginning. The first step in
the process of a new radio system design is to determine base station arrang
ement and a frequency plan, both of which are chiefly dependent on environme
ntal characteristics.
在过去几十年里，无线通信系统得到了巨大的发展。无线系统所受到的制约也与日俱增
。为了寻求高质量、低成本的解决方案，一个新的无线系统必须从刚开始就加以精心设
计。对于一个新的无线系统设计而言，处理的第一个步骤就是确定基站位置以及频率规
划，这两件问题的解决主要取决于环境特性。
One of the most important characteristics of the propagation environment is
the path (propagation) loss. An accurate estimation of the propagation losse
s provides a good basis for a proper selection of base station locations and
a proper determination of the frequency plan. By knowing propagation losses
, one can efficiently determine the field signal strength, signal-to-noise r
atio (SNR), carrier-to-interference (C/I) ratio, etc.
传播环境最重要的特性之一是路径（传播）损耗。传播损耗的精确估计，可以为适当选
择基站位置和适当确定频率规划方案提供很好的基础。通过了解传播损耗，人们可以有
效地确定信号场强、信噪比（SNR）、载干（C/I）比等。
An accurate prediction of the field strength level is a very complex and dif
ficult task. To date, various field strength prediction methods have been pr
oposed in the literature. This article presents an overview of popular predi
ction models and describes some useful algorithms that are based on the auth
ors' experience, to improve their accuracy. It should be noted that in most
cases the models presented predict a local average value (median, mean, slow
fading) which is of particular interest for those system engineers who are
putting radio-systems in operation. Otherwise, models of time dispersion par
ameters, which are very important for device designers, are not considered.
对场强的精确估计是一件很复杂而又困难的任务。到目前为止，文献中提出了多种不同
的场强预测方法。本文对常用的预测模型进行归纳，基于本文作者的经验，介绍了一些
提高其精确度的有用的算法。应当注意在大多数情形下，所提出的模型预测的是本地的
平均值（中值、平均、慢衰落），这些对进行无线系统设计的系统工程师而言更有价值
。另一方面，时域弥散的模型则对设备生产商更有价值，这里不作介绍。
The main propagation mechanisms of the radio signal are described. Definitio
ns and some basic characteristics of the propagation models are given. Macro
cell, microcell, and indoor propagation models are considered in detail in t
he later sections. In these sections, an overview of popular prediction mode
ls is given. Their advantages and disadvantages are discussed, and in additi
on, some practical improvements of these models, along with some new models,
are described.
本文介绍了电波传播的主要传播机理。给出了传播模型的定义及一些基本的特性。在即
后的几节里详细讨论了宏蜂窝、微蜂窝以及室内传播模型。为此，对主要的预测模型进
行了综述，讨论了其优点及不足。此外，介绍了对这些模型的一些实用化改进以及一些
新模型。
Propagation Phenomena
传播现象
Propagation mechanisms are very complex and diverse. First, because of the s
eparation between the receiver and the transmitter, attenuation of the signa
l strength occurs. In addition, the signal propagates by means of diffractio
n, scattering, reflection, transmission, refraction, etc.
传播机制复杂而多变。首先，源于收发信机的分离，信号的强度出现了传播衰减。此外
，信号通过绕射、散射、反射、透射、折射等方式进行传播。
Diffraction occurs when the direct line-of-sight (LoS) propagation between t
he transmitter and the receiver is obstructed by an opaque obstacle whose di
mensions are considerably larger than the signal wavelength. The diffraction
occurs at the obstacle edges where the radio waves are scattered, and as a
result, they are additionally attenuated. The diffraction mechanism allows t
he reception of radio signals when the LoS conditions are not satisfied (NLo
S case), whether in urban or rural environments.
当尺寸远大于信号波长的（对电磁波）不透明的物体遮挡了收发信机之间的直达射线路
径时，会出现绕射现象。绕射出现在物体的边缘，电磁波被散射开来，其结果导致额外
的信号衰减。不管是在城市或农村环境中，绕射机制使得在不满足直达射线的条件下（
无直达射线情形）也可以接收无线信号。
Scattering occurs when the propagation path contains the obstacles whose dim
ensions are comparable to the wavelength. The nature of this phenomenon is s
imilar to the diffraction, except that the radio waves are scattered in a gr
eater number of directions. Of all the mentioned effects, scattering is the
most difficult to be predicted.
当传播路径上有其尺寸可以和信号波长相比拟的（若干）物体时，出现散射现象。这一
现象的特性类似于绕射，只是电磁波向更多的方向上散开。在前所提及的所有现象中，
散射是最难预测的。
Reflection occurs when the radio wave impinges the obstacle whose dimensions
are considerably larger than the wavelength of the incident wave. A reflect
ed wave can either decrease or increase the signal level at the reception po
int. In cases where many reflected waves exist, the received signal level te
nds to be very unstable. This phenomenon is commonly referred to as multipat
h fading, and the signal is often Rayleigh distributed.
当无线电波入射到其尺寸远大于入射波波长的物体表面时，出现反射现象。在接收点处
反射波既可能加强，也可能减弱信号电平。如果存在很多反射波时，接收信号就会变得
不稳定。这一现象通常称为多径衰落，这时的信号服从瑞利（Rayleigh）分布。
Transmission occurs when the radio wave encounters an obstacle that is to so
me extent transparent for the radio waves. This mechanism allows the recepti
on of radio signals inside buildings in cases where the actual transmitter l
ocations are either outdoors or indoors.
当无线电波遇上的物体在某种程度上对无线电波透明时，会出现所谓透射现象。无论实
际的发射机位置在室内还是室外，这一机理使得建筑物内可以接收无线信号。
Refraction is very important in macrocell radio system design. Due to an inc
onstant refractive index of the atmosphere, the radio waves do not propagate
along a straight line, but rather along a curved one. Therefore, the covera
ge area of an actual transmitter is usually larger. However, as a result of
the fluctuations of the atmosphere parameters, the received signal strength
level is fluctuating as well.
折射在宏蜂窝无线系统设计中非常重要。由于大气具有变化的折射率，无线电波不是沿
着直线传播，而是形成一条弯曲的传播路径。因此，实际发射机的覆盖范围通常较大。
另一方面，作为大气参数变化的结果，接收信号电平通常也会起伏不定。
Since there is frequently no LoS between the transmitter and the receiver, t
he received signal is a sum of components that often stem from several previ
ously described phenomena. Therefore, the received signal level is quite var
iable with respect to time and especially with respect to the receiver or tr
ansmitter displacement. Even a displacement of just a fraction of the wavele
ngth can cause the signal level to change by more than 30dB. These fluctuati
ons are known as short-term (or multi-path) fading. On the other hand, the l
ocal average of the signal varies slowly with the displacement. These slow f
luctuations depend mostly on environmental characteristics, and they are kno
wn as long-term fading. Both slow and fast fading are illustrated in Fig. 1.

由于通常在收发信机之间没有直达路径，接收信号通常是若干分量的合成结果，而这些
分量则是源于前所述及的现象中的若干种。因而，接收信号电平通常随时间，特别是随
接收或发射机的位移而快速变化。即便位移量只是若干分之一波长，这样的位移也会导
致接收信号电平30分贝以上的变化。这一波动即所谓的short-term（多径）衰落。另一
方面，信号变化的平均值随位置的变化而缓慢变化。这一慢速的变化主要取决于环境特
性，又称为long-term衰落（衰减）。图1中说明了两类衰落现象。
Since the short-term fading of the received signal is almost impossible to p
redict, all propagation models estimate either the average or median values.
When averaging is performed, the width of the averaging window should be ch
osen carefully. A window that is too narrow results in uncertain averages, w
hile a window that is too wide can wash out the detailed signal changes depe
nding on the local environment. Typically, a suitable window width for appli
cations in the bigger areas is ±20λ[1], while in smaller areas the window
must be narrower.
由于接收信号的多径衰落通常不能加以预测，因此所有传播模型都旨在估计平均值或中
值。当进行平均时，平均窗口的宽度应当加以仔细选择。过窄的窗口会使得平均结果不
可靠，而过宽的窗口则会掩盖本地信号变化的一些细节。典型情况下，在较大的场地(a
rea)下适当的窗口宽度应当在±20λ左右[1]，而在较小的场地内窗口宽度应当变窄。
Propagation Models
传播模型
A propagation model is a set of mathematical expressions, diagrams, and algo
rithms used to represent the radio characteristics of a given environment. G
enerally, the prediction models can be either empirical (also called statist
ical) or theoretical (also called deterministic), or a combination of these
two. While the empirical models are based on measurements, the theoretical m
odels deal with the fundamental principles of radio wave propagation phenome
na.
一个传播模型，是一组用于表示给定环境中无线电传播特性的数学表达式、图表或算法
。一般说来，预测模型可以是经验模型（又称统计模型），也可以是理论模型（也称确
定模型），或是二者的组合。经验模型主要基于测量的结果，而理论模型则处理电波传
播现象的基本原理。
In the empirical models, all environmental influences are implicitly taken i
nto account regardless of whether they can be separately recognized. This is
the main advantage of these models. On the other hand, the accuracy of thes
e models depends not only on the accuracy of the measurements, but also on t
he similarities between the environment to be analyzed and the environment w
here the measurements are carried out. The computational efficiency of these
models is usually satisfying.
在经验模型中，隐含地考虑了所有环境因素的影响，不管它们彼此之间是互相独立的或
是互相影响的。这也是这类模型的主要优点。然而，这些模型的精确程度不仅取决于测
量的精确程度，而且还和所分析的环境与所测试的环境二者间的相似程度有关。这类模
型的计算效率通常能满足要求。
The deterministic models are based on the principles of physics and, due to
that, they can be applied to different environments without affecting the ac
curacy. In practice, their implementation usually requires a huge database o
f environmental characteristics, which is sometimes either impractical or im
possible to obtain. The algorithms used by deterministic models are usually
very complex and lack computational efficiency. For that reason, the impleme
ntation of the deterministic models is commonly restricted to smaller areas
of microcell or indoor environments. Nevertheless, if the deterministic mode
ls are implemented correctly, greater accuracy of the prediction can be expe
cted than in the case of the empirical models.
确定模型基于物理原理，因此，在用于不同环境中是不影响其精确程度。实际上，其实
现通常需要一个很大的环境特性数据库，后者有时是不切实际的或是不可能的。确定模
型所使用的算法通常很复杂且计算效率不高。因为这一原因，确定模型的应用通常局限
于微蜂窝或室内环境等较小的场地范围。不过，一旦确定性模型得以正确应用，可望得
到较之经验模型更为准确的预测结果。
On the basis of the radio environment, the prediction models can be classifi
ed into two main categories, outdoor and indoor propagation models. Further,
in respect of the size of the coverage area, the outdoor propagation models
can be subdivided into two additional classes, macrocell and microcell pred
iction models.
基于无线环境的不同，预测模型可以分为室外和室内传播模型两种情形。而且，根据覆
盖范围大小之别，室外传播模型还可以进一步细分为宏蜂窝和微蜂窝预测模型两种子情
形。
Macrocell Propagation Models
宏蜂窝传播模型
Macrocell design philosophy is based on the assumptions of high radiation ce
nterlines, usually placed above the surroundings; transmitter powers on the
order of several tens of Watts; and large cells whose dimensions are on the
order of several tens of kilometers. Under these assumptions, LoS conditions
are usually not satisfied and the signal from the transmitter to the receiv
er propagates by means of the diffraction and the reflection. Also, for larg
e cells the effects of refraction are very important. All of these factors m
ake the problem of field strength prediction very difficult. For years, a la
rge number of researchers have been struggling with this problem. As a resul
t a large number of models have been proposed. The list includes, but is not
limited to: the Bullington model [2], the model of Okumura et al. [3], the
ITU (CCIR) model [4], the Hata model [5], the "Clearance angle" method [6],
the Polish Administration method [7], the Longley-Rice method [8], the Lee m
odel [9], the EPM-73 method [10], the Deutche Bundest Post (DBP) method [11]
, the Ibrahim-Parsons model [12], the Atefi-Parsons model [13], the Joint Ra
dio Committee (JRC) model [14], the TIREM model [10], the Walfish-Bertoni mo
del [10, 15], the Ikegami method [16, 17], the IRT method [18], the ETF-ANN
Macrocell Model [19], the Ericsson model 9999 [20], etc. In the following te
xt, only a few very popular models are discussed.
宏蜂窝的设计原理基于高辐射中心线的假设，发射机通常高出周围的建筑，其功率在数
十瓦量级，形成其半径可达数十公里量级的大蜂窝。在这样的假设下，通常不能满足直
达射线条件，且从发射机到接收机的信号传播借助于绕射及反射机制。此外，对大的小
区而言折射也很重要。所有这些因素使得场强预测问题变得很困难。在几年的时间里，
大量的研究人员致力于解决这一问题，其结果是出现了大量模型。如果列出一个清单，
则包括Bullington模型[2]、奥村等人的模型[3]、ITU（CCIR）模型[4]、Hata模型[5]、
"透过角"模型[6]、Polish Administration模型[7]、Longley-Rice模型[8]、李氏模型
[9]、EPM-73模型[10]、Deutche Bundest Post模型[11]、Ibrahim-Parsons模型[12]、
Atefi-Parsons模型[13]、Joint Radio Committee (JRC)模型[14]、TIREM模型[10]、W
alfish-Bertoni模型[10,15]、Ikegami模型[16,17]、IRT模型[18]、ETF-ANN宏蜂窝模型
[19]、爱立信模型9999[20]等等，有些甚至还未列出。在以下的内容里，我们只讨论其
中一些最常用的模型。
Model of Okumura et al.
奥村（Okumura）等人的模型
The Okumura et al. method [3] is based on empirical data collected in detail
ed propagation tests over various situations of an irregular terrain and env
ironmental clutter. The results are analyzed statistically and compiled into
diagrams. The basic prediction of the median field strength is obtained for
the quasi-smooth terrain in the urban area. The correction factor for eithe
r an open area or a suburban area should be taken into account. The addition
al correction factors, such as for a rolling hilly terrain, the isolated mou
ntain, mixed land-sea paths, street direction, general slope of the terrain
etc., make the final prediction closer to the actual field strength values.
奥村等人的模型[3]基于经验数据，这些数据源于在各种不规则地形和环境分布下进行的
详细的传播测试。这些结果以统计方法进行分析并合成为图表。在城区准光滑地形下可
以得到中值场强的基本预测结果。在开阔地带或郊区都有可供使用的修正因子。其他的
一些修正因子包括起伏的丘陵地貌、孤立的山峰、混合的陆地海面路径、街道走向、一
般的斜坡地貌等等，这使得最终的预测结果接近于实际的场强值。
In the present engineering practice, the Okumura et al. method is widely use
d. This is a method originally intended for VHF and UHF land-mobile radio sy
stems and involves neither complex computations nor an elaborate theory. Muc
h of its experimental data have been incorporated in the ITU (CCIR) referenc
e curves as well as in other popular models.
在当前的工程实践中，奥村等人的方法得到了广泛的使用。这一方法最初只是拟用于VH
F 和UHF频段的陆地移动无线系统，没有复杂的计算或精细的理论。其中的大多数经验数
据与ITU（CCIR）的参考曲线即其他一些常用模型一致。
However, many authors [9, 21, 23] show certain reserve toward the applicatio
n of the Okumura model. They note that extensive data regarding its performa
nce must be obtained before its use may be advocated. In addition, more care
ful interpretation of the definitions of various parameters needs to be made
. When assessing the values of the model's parameters, the influence of the
subjective factors is not easy to avoid, thus yielding different results for
the same problem.
然而，很多作者[9, 21, 23]表明其对奥村模型的应用持某些保留意见。他们指出在提倡
使用这一模型之前还需要得到更多的关系到其性能的数据。此外，还需要对各种参数进
行更仔细的定义。在评估模型参数的量值时，不可避免各种主观因素的影响，因此对同
一问题会产生不同的结果。

---

时间:  2004-12-18 13:05

作者: mick

In order to make the Okumura technique suitable for computer implementation,
Hata has developed the analytic expressions for the medium path loss for ur
ban, suburban, and open areas [5, 22]. Although these expressions are only a
pproximations and therefore have some limitations, they are almost always us
ed in practice instead of the basic Okumura curves.
为了使奥村这一技术适于计算机处理，Hata对城区、郊区及开阔地带的中值路径损耗提
出了解析表达式[5, 22]。尽管这些表达式只是近似的因而存在某些限制，但在实际中得
到了普遍使用，甚于基本的奥村曲线。
ITU (CCIR) Model
ITU（CCIR）模型
The CCIR [4] method is based on the statistical analysis of a considerable a
mount of experimental data obtained by measurements in many countries. The c
urves for the field strength prediction refer to the kind of rolling irregul
ar terrain found in many parts of Europe and North America, for which a valu
e of parameter Δh, defining the degree of terrain irregularity of 50m, is c
onsidered representative. Parameter Δh is defined as the difference in the
heights exceeded by 10 percent and 90 percent of the terrain over propagatio
n paths in the range of 10km to 50km from the transmitter. To determine the
field strength over any irregular terrain, the attenuation correction factor
dependent upon Δh and given in the form of diagrams should be subtracted f
rom the value read from the reference field strength curves. However, many p
apers ([4] Rep. 239, [8] and [23]) have demonstrated the single parameter Δ
h to be inadequate for precise determination of the attenuation correction f
actor. In addition, the local terrain effects in the region of the receiving
area are in no way taken into account when applying the CCIR method. Finall
y, the field strength reference curves given in [4], Rep. 567 are deduced fr
om the curves given in [4], Rec. 370. The original curves are intended for u
se in planning broadcasting services for the solution of interference proble
ms over a wide area and not for point-to-point communications. Therefore, fo
r rural environments it is not unusual to find the median field strength to
differ by more than 20dB with respect to the predicted value. In urban areas
the error can be even greater. As a consequence, the ITU model is used rare
ly in its basic form. However, due to its simplicity the model is used for f
requency coordination and frequency planning purposes in the border areas (f
or example, between countries).
CCIR[4]方法基于是基于相当多实验数据的统计分析，这些数据是在很多国家实测得到的
。用于场强预测的曲线背景是在欧洲和北美很多地方可以见到的不规则起伏地形，预测
中用一个参数值Δh来定义50米的地形不规则程度，这一数据被理解为具有典型意义。参
数Δh定义为离开发射机10公里到50公里的范围内，高度差超过10%，且90%的地形在传播
路径上。为了确定任意不规则地形上的场强，可以在从场强曲线参考上读出的衰减量中
，减去以图表的形式给出的由Δh所确定的衰减因子。然而，很多文献([4] Rep. 239,
[8] and [23])说明了仅用单一的参数Δh不足以精确地确定衰减修正因子。此外，在应
用CCIR方法时，根本没有考虑接收区域本地地形的影响。最后，[4], Rep. 567中所给出
的场强参考曲线是[4], Rec. 370中所给曲线的推演。而最初的曲线，则是用于提供大范
围广播业务中干扰问题的解决方案而不是针对点对点通信系统。因而在农村地区，中值
场强与预测值相差20分贝以上是常见的现象。在城市地区这一误差甚至会更大。其结果
是ITU很少不加修改就加以使用。然而，因为其简单性，该方法常用于处理边界地区（如
国界）的频率协调和频率规划。
"Clearance Angle" Method
"透过角"方法
The "clearance angle" method [6] is proposed by the European Broadcasting Un
ion (EBU), and adopted by CCIR ([4], Rep. 239). The main ideas of the method
have been to retain the CCIR reference field strength curves given in [4],
Rec. 370, the simplicity of application, and to improve the accuracy by taki
ng into account the terrain effects in the region of the receiving area, sin
ce the latter are not always adequately represented by parameter Δh. These
terrain effects are incorporated through the correction based on a "terrain
clearance angle." This angle should be a representative of those angles in t
he reception area measured between the horizontal line at the receiving ante
nna, and that which clears all obstacles within 16km in the direction of the
transmitter. The correction factor in terms of the "clearance angle" is giv
en in the form of two curves, one for VHF and the other for the UHF band. Th
e curves are derived through an optimization process and are the results of
calculations of the field strength on more than 200 paths in Europe and thei
r comparison to the results obtained by the CCIR method. This correction fac
tor should be added to the field strength level obtained from the CCIR refer
ence curves given in ([4], Rec. 370).
"透过角"方法[6]由欧洲广播联盟（EBU）所提出，并为CCIR所采纳([4], Rep. 239)。这
一方法的主要思路在于保留文献[4] , Rec. 370中给出的CCIR参考场强曲线以简化应用
，并在接收地区范围考虑地形因素的影响以提高其精度，由于后者通常不宜由参数Δh充
分表示，这些地形影响可以用基于"地形透过角"的修正组合为一体。该角度是在接收区
域测得的这样一些角度的典型值，即接收天线处的水平线与发射机方向上16公里范围内
没有障碍物的方向上之间的角度。由"透过角"表示的修正因子以两条修正曲线的形式给
出，分别用于VHF和UHF波段。曲线是通过优化处理而得出的，为此，将欧洲地区200条路
径上的场强结果与CCIR方法所得到的结果进行了比较，在此基础上计算获得。这一修正
因子应叠加在由CCIR参考曲线([4], Rec. 370)所得到的场强电平上。
In its very nature the "clearance angle" model as well as many other models
predict the field level in the rural environment. The effects of urbanizatio
n, forests, overpasses, underpasses, etc., must be taken into account throug
h the appropriate additional correction factors. These corrections, which to
a great extent depend on the specific environment, can be easily determined
through measurements. An example of the "clearance angle" corrections deter
mined for the area of the city of Belgrade (which is a typical Mediterranean
city with a population of two million) is presented in Table 1 [24]. In thi
s case, the eight location types are defined as:
作为其特点，"透过角"模型和其他许多模型一样，可以预估乡村环境中场强电平。对都
市环境、森林、天桥、地下通道等等所产生的影响，可以通过引入修正因子来加以考虑
。这些修正在很大程度上取决于特定环境，可以通过测量较容易地得到确定。表1中给出
的例子，给出了一个贝尔格莱德（一个有两百万人口的典型的地中海城市）城内区域的
"透过角"修正的例子[24]。八类位置类型定义为
A  Dense urban areas (5-10 floor buildings) 密集的城市建筑区域（5-10层楼）
B  Urban areas with high buildings (15-25 floors) that are several hundred m
eters distanced from each other具有高层建筑（15-20层楼）的城市区域，建筑物彼
此间的距离为数百米
C  Suburban areas郊区
D  Residential areas住宅区
E  Rural areas乡村地区
F  Forests and park lands森林和公园
G  Bridges and overpasses桥梁和天桥
H  Tunnels and underpasses less than 50m in length长度在50米内的隧道和地下通
道
表1 "透过角"修正因子
位置类型    城区发射机位置  乡村发射机位置
    K1（dB）    s1（dB）    K2（dB）    s2（dB）
A   8.9 13.9    10.6    14.4
B   4.5 11.4    4.5 9.8
C   6.9 11.4    5.5 14.0
D   4.2 10.3    3.1 12.7
E   +0.6    9.7 0.5 11.4
F   5.8 19.5    13.2    13.5
G   +8.4    11.7    +9.0    12.2
H   12.4    5.5 16.2    7.6
   K-修正，s-误差的标准偏差
It should be noted that the signal in urban environments not only undergoes
additional attenuation, but because of many obstructions, it also fluctuates
more rapidly. As a result, the standard deviation of the prediction error i
s usually greater in urban environments than in rural ones [23, 24].
应当指出的是，城市环境中的信号不仅有额外的衰减，而且由于存在很多障碍物，信号
电平的波动变化也很快。作为结果，在城市中预测结果的标准偏差通常大于乡村环境[2
3，24]。
Based on numerous field strength measurements and laborious statistical proc
essing, the "clearance angle" method achieves quite remarkable prediction ac
curacy [23, 24]. In addition, this model is very operative, not complex, dep
rived of the subjective factors, and exceptionally well suited for computer
implementation. Therefore, many old analog systems (conventional systems, tr
unking systems, etc.) as well as digital mobile radio systems (paging, NMT,
GSM, etc.) have been designed by employing the "clearance angle" model.
基于大量的场强测试和繁杂的统计处理，"透过角"方法可以达到相当可观的预测精度[2
3,24]。此外，这一方法还具有高效、简便、排除主观因素以及特别适合于计算机处理等
优点。因而，很多旧的模拟系统（传统的系统、集群系统等）以及数字移动无线系统（
寻呼、NMT、GSM等）都采用"透过角"模型来进行设计。
Ericsson Model 9999
爱立信模型9999
Model 9999 [20] was developed (and extensively used) by Ericsson engineers a
nd engineers employed in companies supported by Ericsson for the purpose of
designing a cellular system (especially for NMT, GSM, PCS, DCS, etc.). This
prediction model is based on the Okumura-Hata model, and has the form of a v
ery simple analytical expression containing several free parameters. In addi
tion, the model takes into account extra losses due to "knife-edge" diffract
ion over a dominant obstacle and the earth's curvature. Also, Model 9999 req
uires clutter (land usage) database.
模型9999用于设计蜂窝系统（特别是NMT、GSM、PCS、DCS等），是由爱立信所支持的公
司中的工程师所提出，并为他们所广泛使用。这一预测模型基于Okumura-Hata模型，其
形式为包括若干自由参数的一个很简单的解析表达式。此外，这一模型还考虑了在主要
障碍物上的"锋刃"绕射以及地球曲率所引入的额外损耗。还有，9999模型需要地形数据
库。
The model-free parameters and correction factors of each clutter type are de
termined empirically for any specific environment. Due to the simplicity of
the model, its accuracy is very sensitive to the accuracy of measurement dat
a to which the model is adjusted. Usually, prior to the implementation of th
e model, extensive test measurements are carried out in order to collect the
data. It should be noted that Model 9999's main advantage is that it is ver
y fast.
在任意特定环境下，与模型无关的参数以及每种地形的修正因子根据经验来确定。由于
模型很简单，其精确度对调整模型所需的测量数据的精度很敏感。通常在建立模型之前
，需要进行广泛的测量以便于收集数据。应当指出，9999模型的主要优点在于处理速度
快。
Lee Model
李氏模型
W. C. Y. Lee proposed this model in 1982 [9]. In a very short time it became
widely popular among researchers and system engineers (especially among tho
se employed in U.S. companies) mainly because the parameters of the model ca
n be easily adjusted to the local environment by additional field measuremen
ts. By doing so, greater accuracy of the model can be achieved. In addition,
the prediction algorithm is simple and fast. Various radio systems are desi
gned by using this model (AMPS, DAMPS, GSM, IS-95, PCS, etc.).
1982年，W. C. Y. Lee提出这一模型。在很短的时间内，这一模型就在研究人员及系统
工程师中得到了广泛的流行（特别是那些美国公司的雇员）。这主要是因为利用额外的
外场测试结果很容易对模型的参数进行调整。由此，模型可以达到更高的精确度。此外
，预测算法简单快速。利用这一模型可以进行各种各样的无线系统设计（AMPS, DAMPS,
GSM, IS-95, PCS, 等）。
The model consists of two parts. The first part, an area-to-area prediction,
is used to predict a path loss over a general flat terrain without taking i
nto account the particular terrain configuration. Obviously, the area-to-are
a prediction alone is inadequate for hilly regions. The second part of the L
ee model uses the area-to-area prediction as a basis and then develops a poi
nt-to-point prediction, thus resolving the problem. Based on the terrain pro
file database, the point-to-point prediction considers whether LoS condition
s exist or not. In the case of LoS existence, the influence of the reflected
radio waves is carefully examined. On the other hand, when LoS existence is
missing, the obstructions are modeled in the form of "knife-edges" and diff
racted waves are computed.
这一模型由两部分组成。前一部分，区域到区域预测，用于预测一般的平坦地形上的路
径损耗，不需要考虑特定的地形结构。很明显，在多山的地区单独使用区域到区域的预
测是不合适的。李氏模型的第二部分以区域到区域的预测为基础，提出了一种点到点的
预测，从而来解决有关问题。基于地形地貌数据库，点到点的预测考虑是否存在直达射
线。存在直达射线时，应仔细检查反射的无线电波的影响。另一方面，当不存在直达射
线时，则以"刃状边缘"作为障碍物的模型并进行绕射波的计算。
The basic area-to-area model can be expressed in the following form:
基本的区域到区域模型可以如下进行表述：
(1)
where Pr is the signal power in W at distance r from the transmitter; for th
e signal frequency f, Pr0 is the signal power at the point of interception a
t distance r0 from the transmitter; for the reference frequency f0, γ denot
es path-loss slope, n denotes frequency dependence, while γ0 is an adjustme
nt factor for antenna heights, transmitter power, and antenna gains. The bas
ic area-to-area model can be extended to a more general case when the radio
wave propagates along several different environments. In that case, the path
loss slopes γi as well as the boundaries of each environment must be known
. The parameters of Eq. (1) depending on the environment are Pr0 and γ. For
several specific environments, mostly in the United States, Lee found value
s of Pr0 and γ [9]. It should be noted that the estimation accuracy mainly
depends on these parameters and thus they must be as precise as possible. Th
e parameters Pr0 and γ for the particular environment can be easily determi
ned through measurements of signal strength. The results of the extensive pr
opagation study carried out in the area of Belgrade are shown in Table 2 [25
]. The results for this urban environment are different from parameters obta
ined by Lee himself. This can be easily explained by the fact that the struc
ture of a Mediterranean city is completely different from the structure of a
typical American town.
设Pr是离开发射机距离为r处单位为瓦（W）的信号频率为f的信号功率，Pr0是离开发射
机距离为r0处测试点处的信号频率为f0的参考信号功率，g 表示路径损耗因子，n表示频
率相关因子，而g0为天线高度、发射机功率、天线增益等的调整因子。最基本的区域到
区域的模型可以扩展为包括无线电波沿若干不同的传播路径传播在内的更为一般性的模
型。在这一情形下，路径损耗因子 gi以及每一环境（路径）的边界都必须已知。方程（
1）中取决于环境的参数是Pr0和g 。对若干特定的环境，主要分布在美国，李确定了参
数Pr0和g 的取值[9]。应当指出的是，估计的精度主要取决于这些参数，因而这些参数
应当尽可能准确。特定环境下参数Pr0和g的取值可以通过对信号强度的测量而方便地加
以确定。对贝尔格莱德地区进行的广泛的传播研究的结果参见表2中[25]。这一城区环境
的结果不同于李氏所得出的参数。对此，可以用这样的事实来加以解释：一个地中海城
市的结构完全不同于一个典型的美国乡镇的结构。
表2 李氏模型的贡献
环境类型    P0（dBm）   g（dB/dec）
农村地区    -57.0   40.3
森林或公园  -57.0   44.5
住宅区  -57.0   47.0
郊区    -59.2   47.3
城区（建筑高度：高达四层）  -61.5   35.4
远处有高层建筑的城区（1525层）  -61.5   37.3
密集的城区（建筑高度：高达四层）    -61.5   55.8
密集的城区（建筑高度：六层以上）    -61.5   56.9
PT =10 W, f = 900 MHz, r0 =1英里（mile），gr=6 dBd, gr = 0 dBd, hT =30 m, hr
=3 m
COST 231-Walfisch-Ikegami Model
COST 231-Walfisch-Ikegami模型
The COST 231-Walfisch-Ikegami model (COST 231-WI) [26] has been used extensi
vely in typical suburban and urban environments where the building heights a
re quasi-uniform. It should be noted that the designers of public mobile rad
io systems (e.g., GSM, PCS, DECT, DCS, etc.) often use this model.
COST 231-Walfisch-Ikegami 模型 (COST 231-WI) [26]广泛应用于其建筑物高度相差不
大的典型的郊区及城区环境。值得注意的是，公共移动无线系统(例如, GSM, PCS, DEC
T, DCS,等)的设计人员经常使用这一模型。
The model utilizes the theoretical Walfisch-Bertoni model [15] to obtain mul
tiple screen forward diffraction loss for high base station antenna heights,
whereas it uses measurement-based data for low base station antenna heights
. This model also takes into account free-space loss, loss due to diffractio
n down to the street, and the street orientation factor.
这一模型采用Walfisch-Bertoni理论模型[15]来得到较高基站天线高度情形下的多重前
向绕射损耗，而对较低的基站天线高度则采用基于测量的数据。这一模型同时考虑了自
由空间传输损耗、到达街道的绕射损耗，以及街道走向等因素。
Steep transitions of path loss occur when the base station antenna height is
close to the height of the rooftops of the buildings in its vicinity. There
fore, the height accuracy of the base station antenna is especially signific
ant if large prediction errors are to be avoided. Moreover, the performance
of the Walfisch-Ikegami model is poor when the base station antenna height i
s significantly lower than the heights of the rooftops of adjacent buildings
.
当基站天线高度接近其附近的建筑物屋顶高度时，将会出现路径损耗快速变化的过渡区
（损耗不连续处）。因此，如果要避免大预测误差的话，基站天线高度的精确性特别重
要。此外，当基站天线的高度明显低于其附近建筑物屋顶高度时，Walfisch-Ikegami模
型的性能很差。
It was claimed, as the expected accuracy of the model, that the mean error i
s in the range of ±3dB and the standard deviation is about 4-8dB in the cas
e when the base station antenna height is several meters above the highest r
ooftops of adjacent buildings within a radius of approximately 150m. However
, recently it was found that the loss expression for the diffraction from th
e last rooftop to the street in the COST 231-WI model is over 8dB more optim
istic than it is supposed to be [27].
据说正如该模型所预期的精度，当基站天线高出其周围150米半径范围内相邻建筑物最高
屋顶高度数米时，其均值误差在±3dB的范围内且标准偏差为4-8dB。然而，最近发现在
COST 231-WI模型中从最后的屋顶到街道的绕射损耗，实际较之模型中情况要有8dB的改
善[27]。
ETF-Artificial Neural Networks Macrocell Model
ETF-人工神经网络宏蜂窝模型
Recently, several prediction models utilizing artificial neural networks (AN
N) [19, 28] have been proposed. The main intention of the work presented in
[19] was to form a good prediction model, i.e., the model that can ensure hi
gh accuracy (exceeding the accuracy of the most popular models) in real time
that uses just ordinary (easily obtainable) databases.
最近，提出了若干采用人工神经网络(ANN，artificial neural networks)预测模型[19
, 28]。文献[19]中所介绍工作其主要目的是构成一个好的预测模型，即仅仅使用普通的
（容易获得的）数据库，这一模型可以保证实时的高精确度（超过大多数流行模型的精
确度）。
The ANN model, proposed in [19], is based on a very popular feed-forward neu
ral network architecture (precisely, multilayer perceptron) [29]. Feed-forwa
rd neural networks with sigmodial activation functions have demonstrated ver
y good performance in solving problems with mild nonlinearities on the set o
f noisy data. That case fully corresponds to the problem of the field streng
th prediction. The data obtained by measurements are always noisy. Another k
ey feature of neural networks is the intrinsic parallelism allowing for a fa
st evaluation of solutions. The process of learning may last for a couple of
hours, but the process of field strength prediction is fast.
文献[19]中提出的ANN模型基于一种非常流行的正馈神经网络结构（更准确的说法是多层
感知）[29]。具有sigmodial激活功能的正馈神经网络求解问题时的优越性表现在，适用
于对含噪声数据集上进行的适度非线性处理。这一清形完全可以与场强预测问题对应，
因为由测量获得的数据总是含有噪声的。人工神经网络的另一个主要特点是其内在的并
行处理机制可以实现快速求解。学习的过程可能会持续一两个小时，但场强预测的过程
会很快。
The proposed neural network has three groups of inputs. The first group cons
ists of an input only and it is the normalized distance from the transmitter
to the receiver. The second group of inputs (4 inputs) is based on the terr
ain profile analysis. These inputs are: 1) portion through the terrain; 2) a
nd 3) modified "clearance angle" factors for both the transmitter and the re
ceiver sites, respectively; and 4) the rolling factor. The third group of in
put parameters is based on the land category analysis along the straight lin
e drawn between the transmitter and the receiver. There is a single input fo
r each defined land use category. The network has one output and it is a nor
malized electric field level. The implementation of the proposed ANN model r
equires two databases. The first is the standard digital terrain elevation d
atabase; the other is the ground cover (i.e., land usage or "clutter") datab
ase.
这一神经网络有三组输入。第一组仅有一个输入即发射机和接收机之间归一化的距离。
第二组输入是基于地形地貌分析的四个输入。这些输入包括：1)通过部分的地形; 2) 及
3) 分别对应于发射机和接收机位置的修正的"透过角"因子; 以及 4) 起伏因子。第三组
输入参数基于连接发射机和接收机的一条直线上的沿线陆地分类分析。对每一种定义的
陆地使用分类取一个单一的输入。网络只有一个输出即归一化电场的电平值。ANN模型的
实现需要有两个数据库。第一个是标准数字地形海拔数据库，另一个是地面覆盖（地貌
）物（即陆地使用情况或"混乱"情况）数据库。
In comparison to other popular prediction models, the ANN model demonstrated
very good performance. The effects of urbanization are considered more subt
ly in the proposed model than in standard empirical models providing greater
accuracy. On the other hand, the ANN model is not computationally as extens
ive as deterministic models.
相对于其他流行的预测模型，ANN模型具有很好的性能。这一模型较之标准的经验模型，
将城市化的影响进行了巧妙的处理，并得到更精确的结果。另一方面，ANN模型较之确定
性的模型减少了计算的开销。
The ANN model has been realized and used in 450MHz and 900MHz frequency band
s for the purpose of TETRA and GSM system design, respectively.
针对TETRA和GSM系统设计，ANN模型分别在450MHz和900MHz频段得到了实现和应用。
Microcell Propagation Models
微蜂窝传播模型
A microcell is a relatively small outdoor area such as a street with the bas
e station antenna below the rooftops of the surrounding buildings. The cover
age area is smaller compared to macrocells and it is shaped by surrounding b
uildings. A microcell enables an efficient use of the limited frequency spec
trum and it provides a cheaper infrastructure. The main assumptions are rela
tively short radio paths (on the order of 200m to 1000m), low base station a
ntennas (on the order of 3m to 10m), and low transmitting powers (on the ord
er of 10mW to 1W). Today, microcells are very often used in IS-95, PCS, DCS,
GSM, DECT, etc.
微蜂窝是一个相对较小的室外区域如街道，其基站天线低于周围建筑物屋顶。其覆盖区
域较宏蜂窝要小，且主要由周围确定覆盖形状。微蜂窝有助于有限频率资源的有效利用
，且其基础设施投入较少。主要的设计指标是较短的无线电路径(在200米到1000米的量
级)，较低的基站天线高度(在3米到10米的量级)，以及较低的发射功率(在10毫瓦到1瓦
的量级)。今天，微蜂窝已经广泛应用于IS-95, PCS, DCS, GSM, DECT等系统中。
There are many prediction models for a microcell situation. In this article,
the authors present a review of some interesting models that are extensivel
y used in the process of designing previously mentioned radio systems.
微蜂窝情形下有很多预测模型。本文中，作者对广泛应用于前述无线系统设计过程中的
一些有趣的模型加以综述。
Empirical Models
经验模型
The models proposed in [30, 31] describe the measured signal level along the
line-of-sight path. According to these models, the road-guided waves are ex
pected to exist only for short ranges. This situation can be described by tw
o distinct path-loss slopes and a break point. The break point is the distan
ce from the base station that is equal to the maximum distance that has the
I Fresnnel zone clear. The break point can be used to define the size of a m
icrocell because the signal level decreases more rapidly when the distance i
ncreases after the break point.
文献[30, 31]所提出的模型描述了沿直达射线路径上测量的信号电平。根据这些模型，
沿道路引导的电磁波预期只在较短的范围内存在。这一情形可以解释为两段明显不同的
路径损耗斜率和一个明显的断点。断点离开基站的距离等于第一菲涅尔区的最大距离。
由于断点后的信号电平随距离增大而降低的速度更快，因此，断点可以用于确定微蜂窝
的尺寸。
The form of the proposed propagation models is given by:
传播模型的形式可以表示为
                           (2)
where S is the signal level in dBμV/m, d is a distance from the transmittin
g antenna (m), a is the basic attenuation rate for short distances, b is the
additional attenuation rate coefficient for the distance greater than 100 t
o 200m, g is the distance to the break point, and c is a scaleable factor. T
he expression is valid for 5 - 20m antenna heights and 200m - 1km distances.
This model, whose coefficients are relatively independent, has two boundary
cases:
这里，S是以dBμV/m为单位的信号电平，d是离开发射天线的距离（米），a是短距离的
基本衰减速率，b是大于100到200米时附加的衰减速率因子，g是到断点的距离，且c是可
缩放的因子。这一表达式适用于天线高度为5-20米，传输距离为200米到1公里的情形。
这一模型其系数相对独立，并有两种边界情形：
l   At distances less than the break point, the form of the propagation model
is:
l   当距离小于断点距离时，传播模型可表示为
                                  (3)
l   At distances greater than the break point, the form of the propagation mod
el is:
l   当传播距离大于断点距离时，传播模型可表示为
                           (4)
In addition, the signal around the corner decreases by 20 - 25dB in a short
transition distance of only several tens of meters.
此外，街道拐角处数十米距离上有一个过渡区，信号电平降低可达20-25分贝。
It was shown that for the same conditions, the results of the proposed model
s [30, 31] for a microcell situation are better than those of the normal lin
ear regression and the Okumura model.
同样条件下的有关结果表明，在微蜂窝情形下文献[30, 31]所提模型的预测结果优于其
他标准线性衰退及奥村模型的结果。
Two-Ray Model
两射线模型
Numerous propagation models for microcells are based on a ray-optic theory.
In comparison with the case of macrocells, the prediction of microcell cover
age based on the ray-model is more accurate. One of the elementary models is
the two-ray model. The two-ray model [32] is used for modeling of the LoS r
adio channel and it is described in Fig. 2a.
许多适用于微蜂窝的传播模型都是基于射线光学理论。与宏蜂窝的情形相比，基于射线
模型的微蜂窝覆盖预测更为精确。其中一个基本的模型即所谓两射线模型[32]。这种模
型用于对LoS无线信道建模，并如图2a中所示。
The transmitting antenna of height h1 and the receiving antenna of height h2
are placed at distance d from each other. The received signal Pr for isotro
pic antennas, obtained by summing the contribution from each ray, can be exp
ressed as:
发射天线高度为h1，接收天线高度为h2，且彼此间的距离为d。各向同性（全向）天线的
接收功率Pr，可以由叠加各条射线的贡献而得到，即可表示为：
(5)
where Pt is the transmitter power, r1 is the direct distance from the transm
itter to the receiver, r2 is the distance through reflection on the ground,
and Γ(α) is the reflection coefficient depending on the angle of incidence
α and the polarization.
这里，Pt为发射功率，r1为从发射机到接收机的直达射线路径长度，r2为经过地面反射
后的路径长度，且G(a)为与入射角a和极化有关的发射系数。
The reflection coefficient is given by:
反射系数可以表示为：
(6)
where θ = 90°-α and a = 1/ε or 1 for vertical or horizontal polarization
, respectively. εr is a relative dielectric constant of the ground.
这里，q=90°-a，且对应于垂直及水平极化分别有a=1/e或1。er是地面的相对介电常数
。
In Fig. 2b the received power given by Eq. (5) is shown as a function of the
distance for the cases of horizontal and vertical polarizations as well as
for the case assuming Γ(θ) = -1. For large distances α is small, and Γ(θ
) is approximately equal to -1. For short distances, the value of Γ(θ) dec
reases and it can even be zero for vertical polarization.
在图2b中由(5)式给出的接收功率表示为距离的函数，其中，水平及垂直极化情形下均假
定G( q)=-1。当距离较大时，a较小，因此G( q)近似为-1。而当距离较短时，G( q)取值
将会减小，在垂直极化情形下甚至为零。
Also, there are more complex models based on the ray-optic theory. The four-
ray model consists of a direct ray, ground-reflected ray, and two rays refle
cted by buildings. The six-ray model, besides the direct and the ground-refl
ected ray, takes four rays reflected by the building walls along the street.
If a model considers a larger number of rays, the prediction tends to be mo
re accurate, but the computational time is significantly increased. The prob
lem deserving special attention is that of the corner diffraction. Two popul
ar models considering this effect are the GTD (Geometrical Theory of Diffrac
tion) model [33], and the UTD (Uniform Theory of Diffraction) model [34].
此外，还有更多基于射线光学的理论的更复杂的模型。四射线模型包括直达射线、地面
反射线以及两条建筑物上的反射线。六射线模型除直达射线和地面反射线外，还有来自
沿街道建筑物墙壁反射的四条射线。如果模型中考虑了大量数目的射线，则预测结果将
更为精确，然而计算时间将会显著增长。应受到特别重视的是拐角绕射。考虑这一问题
的两个常用模型是几何绕射理论(GTD, Geometrical Theory of Diffraction)模型[33]
和一致性绕射理论(UTD,Uniform Theory of Diffraction)模型[34]。
Models Based on UTD and Multiple Image Theory
基于UTD和多镜像理论的模型
One of the proposed models is quasi three-dimensional UTD propagation model
[35], which functions well for microcellular applications. A multiple image
concept and generalized Fermat's principle are used to describe the multiple
reflections and diffractions. It is assumed that the building walls are muc
h higher than the transmitter height so that the diffraction from the roofto
ps can be neglected. The model considers various line-of-sight propagation p
aths and, also, non-line-of-sight paths. The propagation paths taking a larg
e number of corners and building walls are not necessarily coplanar. This mo
del includes multiple reflections between wall-to-wall, wall-to-ground, grou
nd-to-wall, the diffraction from the corners of buildings, and, also subsequ
ent reflections from such diffracted signals. The relative contributions of
the diffraction and reflection components to the total received signal along
a side street depend on the parameters such as the widths of the main stree
t, side streets, parallel streets, the distance from the transmitter to the
junction, the reflectivity of the surfaces, etc.
一种建议的准三维UTD传播模型[35]，能够很好地应用于微蜂窝。多镜像的概念和一般化
的费马原理可以用于描述多个反射及绕射。假定建筑物墙壁远高于发射机高度，使得屋
顶的绕射可以忽略不计。这一模型考虑了各种直达射线

---

时间:  2004-12-18 13:05

作者: mick

diffraction causes a loss in signal strength, the value of j will depend on
the values of σ and εr of the walls and ground surfaces as well as the geo
metry of the environment.
UTD模式一次只考虑一条射线。自然地，在特定位置Rx处的接收信号源于很多射线的贡献
。UTD方法将所有反射及绕射射线的贡献进行矢量合成。一般说来，可以到达接收机的总
数为j个的墙壁反射来源于主街道、人行道、平行街道以及最重要的地面反射，也许会有
（也许没有）连接处建筑物拐角产生的绕射。这等价于包括了多个发射机的镜像。由于
每一次反射或绕射都会导致信号强度的降低，j的具体数值取决于墙壁或地表面的s及er
取值，以及环境的几何特点。
Lee Microcell Model
李氏微蜂窝模型
The Lee model for predicting the electric field in microcells [9] assumes th
at there is a high correlation between the signal attenuation and the total
depth of building blocks along the radio path. This assumption is not entire
ly true because the signal received at the mobile unit comes from the multip
ath reflected waves and not from the waves penetrating through the buildings
. However, according to the assumption, if the building blocks are larger, t
he signal attenuation is higher. An aerial photograph can be used to calcula
te the proportional length of a direct wave path being attenuated by the bui
lding blocks. The line-of-sight signal reception curve Plos is determined fr
om the measurement data along the streets in an open line-of-sight condition
. The additional signal attenuation αB curve due to the portion of building
blocks over the direct path can be obtained in the following way:
预测微蜂窝内电场的李氏模型[9]假定无线电路径上信号的衰减与总的建筑物遮挡高度之
间有较高的相关性。这一假定不完全真实，因为移动单元接收到的信号来自于多径反射
，而不是穿透建筑物而来的直射波。然而按照这一假定，如果建筑物的遮挡很大则信号
衰减增大。可以利用一张航空照片来计算直达波被建筑物遮挡衰减的比例长度。在开放
的直达射线条件下，可以通过沿街道进行测量的数据来确定直达射线信号接收曲线Plos
。在此基础上，可以依照下面的方法进而确定由于直达路径上建筑物遮挡所造成的额外
信号衰减aB曲线
l   Calculate the total blockage length by adding the individual building bloc
ks.
l   通过增加单个建筑物遮挡来计算总的遮挡长度
l   Measure the signal strength Plos for a line-of-sight condition.
l   在直达射线条件下测量信号强度Plos
l   Measure the signal strength Pnlos for a non-line-of-sight condition.
l   在非直达射线条件下测量信号强度Pnlos
l   If the signal level at a particular point is Pnlos, the distance from the
base to the mobile unit is dA, and B is the blockage length between the tran
smitter and the receiver, then the value of αB for a blockage B can be expr
essed as:
l   设一特定点处的信号电平为Pnlos，基站到移动单元的距离为dA，B为发射机到接收机
之间的遮挡长度，则遮挡B所对应的aB值可以表示为
aB(B) = Plos(d = dA) - Pnlos                             (7)
The additional signal attenuation αB curve based on the building blockage a
nd the line-of-sight measured path loss are shown in Fig. 4. These curves ar
e found experimentally [9]. A series of measurements have been done for diff
erent antenna heights in LoS conditions along many streets, and it is observ
ed that the antenna height gain for different antenna heights is 30dB/dec.
附加的信号衰减aB曲线基于建筑物遮挡和图4中所表示的直达射线路径衰减的测量值。这
些曲线是经验基础上建立的[9]。沿很多街道在直达射线的条件下针对不同的天线高度进
行了一系列的测量，从测量结果中观察到不同天线高度下天线的高度增益可达30 dB/de
c。
In conclusion, in a microcell prediction model, two curves Plos and αB are
used to predict the received signal strength. Therefore, the microcell predi
ction model is given by:
作为其结论，在微蜂窝预测模型下，用两条曲线Plos和aB来预测接收信号强度。因此蜂
窝预测模型可以表示为
Pr = Plos - aB                                  (8)
The original Lee model exhibits large errors in the following situations:
在下列情况下，最初的李氏模型表现出较大的误差：
l   When the prediction point is in the main street, but there is no direct Lo
S path.
l   预测点在主街道上，但该位置没有直接的直达射线路径
l   When the prediction point is in a side street near an intersection and lar
ge building blocks exist between the point of prediction and the transmitter
(the case when the side street and the transmitter location are on the same
side of the main street).
l   人行道上的预测点靠近街口，且发射机与预测点之间存在很大的建筑物遮挡（及发射
机和人行道在主街道的同一侧）
The accuracy of the model can be significantly improved by introducing speci
fic corrections based on the arrangement of the streets and their types [36,
37]. There are significant differences in the propagation of radio waves in
different types of streets. (For example, a main street under LoS condition
s, a main street under NLoS conditions, a narrow side street, a wide side st
reet, and a street parallel to the main). After these corrections are added
to the model, the signal level in side streets and in the main street under
NLoS conditions is given by:
通过引入基于街道走向及街道类型的特殊修正后[36, 37]，模型的精度可以得到重要的
改善。在不同类型的街道上无线电波的传播有着本质的区别。（例如，有直达射线的主
街道、不符合直达射线跳进的主街道、较窄的人行道、较宽的人行道以及平行于主街道
的道路等）。当将这些修正加入到原模型中时，人行道和主街道上的非直达射线条件下
的信号电平可表示为
Pnlos[dB] = Plos(LoS-distance) - ast(NLoS-distance)                      (9)

where Pnlos is the estimated signal level in the street under NLoS propagati
on conditions, Plos is the signal level on the LoS path at the intersection
of the main and side street (at a LoS-distance from the transmitter), and α
st is the correction of the signal level in the side street at the NLoS-dist
ance from the intersection.
The results of the prediction in the side street based on the original Lee m
odel and the improved model are shown in Fig. 5 and Fig. 6, respectively. On
the figures, the black color represents the buildings while different shade
s of blue denote different signal levels. In the first part of the street, t
he signal level obtained by the improved model is greater than the signal le
vel obtained by the primary Lee model for the same distance from the interse
ction. The estimated signal levels shown in Fig. 6 correspond more adequatel
y to the real situation. Therefore, the improved model is more precise. At t
he same time, the algorithm is not overly complex.
这里，Pnlos是非直达射线传播条件下街道上的估计电平，Plos是主街道和人行道交汇处
直达射线路径下的信号电平（在离开发射机的直达射线距离上），且ast为离开交汇处非
直达射线距离上人行道处信号电平的修正量。图5及图6中分别给出了基于原始的李氏模
型及改进后的李氏模型下，沿人行道的预测结果。图中，黑色表示了建筑物，而不同深
度的蓝色则表示了不同的信号电平。在街道的第一段，离开交汇处相同距离时改进后模
型所得信号电平高于原始李氏模型结果。图6中表示的估计信号电平与真实情形更为接近
。因此，改进的模型更为精确。同时，算法复杂度没有明显增大。
Indoor Propagation Models
室内传播模型
At first glance, the field strength prediction in indoor environments seems
be to easier than the outdoor prediction. However, measurements [38] show th
at the field strength dynamics can be very high (over 80dB). Also, experimen
tal autocorrelation has shown that a separation of 0.4λ is required for a c
orrelation coefficient below 0.2 between two adjacent samples (Fig. 7) [38].
The same parameter for the outdoor environment is 0.8λ [9]. This differenc
e can be explained by the fact that at a specific location, the electric fie
ld of the indoor environment is formed by a much larger number of indirect c
omponents than in the case of the outdoor environment. Therefore, the indoor
signal level is more fluctuating than the outdoor signal level, and thus it
is more difficult to predict.
初次接触，似乎室内环境的场强预测易于室外预测。然而，测试结果表明[38]场强的动
态变化范围很大(超过80dB)。此外，自相关性的实验结果表明[38]：为了使两个相邻样
本之间的相关系数低于0.2，空间间隔应达0.4l（图7）。而在室外环境下同一的参数则
为0.8l[9]。这一差别可用这样的事实加以解释，即同室外环境的情形相比，在一个特定
位置处室内环境下的电场由更多的非直达射线分量合成。因此，室内信号电平较之室外
信号电平更容易上下变化，因而也就更难于预测。
The problem of the indoor field level prediction can be considered statistic
ally or theoretically. While almost all statistical (empirical) models are b
ased on the same general model, there are several distinguished theoretical
models of which ray-tracing models and Finite-Difference Time-Domain (FDTD)
models are the most popular. Some important disadvantages of both empirical
and theoretical models can be overcome by an appropriate artificial neural n
etwork (ANN) model.
室内场强电平的预测问题可以用统计方法或理论方法加以处理。与几乎所有的统计（经
验）模型均基于相同的通用模型不同，理论模型则有若干种彼此区别甚大的模型，且其
中最常用的则是射线追踪模型及时域有限差分模型。而当采用适当的人工神经网络模型
时，经验模型和理论模型中的一些缺点可以得到克服。
The general idea of each of the presented models can be easily applied to an
y specific frequency band. However, the 1.8-2GHz frequency band is of partic
ular importance because the major indoor radio systems operate today in this
band (DECT, PACS, PHS, etc.)
每一种现有模型的基本思路可以方便地应用到任意特定频段。然而，1.8-2GHz频段具有
特别的重要性，因为当今大多数室内无线电系统(DECT, PACS, PHS等)都工作在该频段。

Empirical Models
经验模型
The general empirical model can be expressed as [30]:
一般的经验模型可以表示为[30]：
PL(d) = PL(d0)+10·n·log(d/d0)+Xs                         (10)
where PL(d) is the path loss in dB at distance d, PL(d0) is the known path l
oss at the reference distance d0 (usually d0 = 1m), n denotes the exponent d
epending on the propagation environment, and Xσ is the variable representin
g uncertainty of the model. Based on this general formulation many empirical
models have been derived [39-44].
这里，PL(d)为距离d处以dB为单位的路径损耗，在参考点d0处（通常d0= 1m）已知的路
径损耗，n为由传播环境决定的指数，且Xσ表示模型不确定性的变量。
These models are simple, efficient, and suitable for computer implementation
. During implementation, the environmental database is unnecessary. Therefor
e, there is no requirement for investing time and resources in surveying bui
lding layouts. Due to model simplicity, great accuracy could not be expected
. The main parameter n is very sensitive to the propagation environment, i.e
., the type of construction material, type of interior, location within buil
ding, etc. The values of n range from 1.2 (waveguide effect) to 6. In additi
on, the value of n depends on the way the statistical analysis on measuremen
t data is performed.
这些模型简单有效，且便于计算机实现。在实现过程中，环境数据库是不必要的。因此
，在测量建筑物内环境分布上不需要投入时间和资源。由于模型简单，就不可能达到很
高的精确程度。主要的参数n对传播环境非常敏感，包括建筑材料的类型、内部的布置以
及建筑物内的位置等。n的变化范围可以从1.2 (波导效应)到6。此外，n的取值还与对测
量数据所采取的统计方式有关。
Ray-Tracing Models
射线追踪模型
The ray-tracing algorithm [33, 34, 41] calculates all possible signal paths
from the transmitter to the receiver. In basic ray-tracing models, the predi
ction is based on the calculations of free-space transmissions and reflectio
ns from the walls. More complex ray-tracing algorithms include the mechanism
of diffraction, diffuse wall scattering, and transmission through various m
aterials. In the end, the signal level at any specific location is obtained
as a sum of the components of all paths between the transmitter and the rece
iver. In addition to the propagation losses, the time dispersion of the sign
al can be successfully predicted by the ray-tracing models.
射线追踪算法[33, 34, 41]计算所有可能的从发射机到接收机的路径。作为基本的射线
追踪模型，预测基于对自由空间传输和墙壁反射进行计算。更复杂的射线追踪算法包括
绕射机制、粗糙墙壁散射，以及各种材料的投射。最后，任意特定位置处的信号电平可
以通过叠加收发信机之间所有路径分量而成。除路径损耗之外，利用射线追踪模型可以
成功地预测信号的时间弥散。
Today, the ray-tracing models belong to a group of the most accurate field s
trength prediction models. However, they require a very detailed layout of t
he area to be analyzed. The accuracy of the model depends on the accuracy an
d complexity of the area layout database. On the other hand, the implementat
ion of these models requires extensive computational resources. Computationa
l time depends exponentially on the details included in the layout of the ar
ea. Therefore, the computational time of a small area with plenty of details
can be greater than that of a big area that is relatively poor in details.
今天，射线追踪模型属于一组最精确的场强预测模型之一。然而，这一方法需要有所分
析区域内详细的布局。模型的精确性取决于区域布局数据库的精确度和复杂性。另一方
面，这些模型的实现需要相当大的计算开销。计算时间对区域布局中的细节呈指数性的
依赖关系。因此，有丰富细节内容的较小区域所消耗的计算时间，会大于相对较少细节
的较大区域消耗的计算时间。
Ray-tracing algorithms can also be used for signal level prediction in outdo
or environments, but for relatively smaller areas.
射线追踪算法同样可以适用于室外环境的信号电平预测，但只适用于相对较小的区域。

Finite-Difference Time-Domain (FDTD) Models
时域有限差分模型
Radio propagation characteristics can be derived solving directly Maxwell's
equations of electromagnetic wave propagation. The FDTD method is probably t
he most popular method for a numerical solution of Maxwell's equations [41].
In this method, Maxwell's equations are approximated by a set of finite-dif
ference equations. Prior to calculations, it is necessary to define a specif
ic grid (regular or irregular) over the area of interest. After appropriate
initial conditions are defined, the FDTD algorithm employs the central diffe
rences to approximate both spatial and temporal derivates. At the nodes of t
he grid, the solutions are determined iteratively. In this way, Maxwell's eq
uations are solved directly.
无线电传播特性可以直接求解电磁波传播所满足的麦克斯韦方程而得到。在麦克斯韦方
程的众多数值解法中，FDTD可能算是最流行的一种[41]。在这一方法中，麦克斯韦方程
被近似化为一组有限差分方程。在进行计算之前，有必要对所分析的区域定义一种特定
的网格（规则或不规则）。在定义了适当的初始条件之后，FDTD算法采用中心差分来近
似空域和时域的微分。在网格的节点上，通过迭代的方式进行求解。这样，可以实现对
麦克斯韦方程的直接求解。
Similarly to the ray-tracing model, the FDTD models are very computationally
demanding. The computational time depends proportionally on the size of the
area to be analyzed, but not significantly on the details involved. However
, the number of nodes of the grid is exponentially related to the size of th
e area and the frequency of operation. The accuracy of the FDTD model is com
parable to that of the ray-tracing models. The prediction is as accurate as
the area layout database.
与射线追踪模型相类似，FDTD模型需要很大的计算开销。计算的时间与所分析区域的尺
寸成比例，而不是由区域内的细节所决定。然而，网格节点的数目随区域尺寸及工作频
率变化呈指数变化。FDTD模型的精确性与射线追踪模型相仿。精确程度与区域布局数据
库的精确程度相关。
Due to computational complexity, FDTD models are suitable only for field pre
diction tasks in small areas. For large areas, ray-tracing models are more s
uitable.
由于其计算的复杂度，FDTD模型只是用于小区域范围内的场强预测。对较大区域，则射
线追踪方法更为适用。
ETF-Artificial Neural Network (ANN) Model
ETF-人工神经网络模型
The main problem presented by empirical models is their unsatisfactory accur
acy. On the other hand, the theoretical models lack computational efficiency
. A compromise can be made by the artificial neural network model [38, 45].
适用经验模型的主要问题是其精确程度不能令人满意。另一方面，理论模型计算效率不
高。采用人工神经网络模型则可以进行折衷[38, 45]。
Similar to the case of the ANN macrocell model, this model is based on multi
layer perceptron feedforward neural networks. The implementation of an ANN m
odel requires a database of the floor plan in which all particular locations
are classified into several environmental categories, for example, wall, co
rridor, classroom, window, etc. The easiest way to do this is to make a colo
r picture over the scanned floor plan.
与ANN宏蜂窝模型的情形相类似，这一模型基于多层感知正馈神经网络。ANN模型的实现
需要有一个楼层布局数据库，其中所有特定位置都应当归于若干种环境类别，例如墙壁
、走廊、教室、窗户等等。做这一工作的最简单的方式是在扫描的楼层布局图上加以各
种色彩表示。
The ANN model proposed in [38, 45] has the form of the multilayer perceptron
with three hidden layers. There are several inputs based on the number of p
reviously defined environmental categories. One of the inputs is the normali
zed distance from the transmitter to the receiver. The remaining inputs are
based on the analysis of the straight line drawn between the transmitter and
the receiver with respect to environmental categories, e.g, how many doors,
what percentage of the line passes through the classrooms, etc. The model h
as an output that is a normalized field level.
文献[38, 45]中所介绍的ANN模型采用含有三个隐含层的多层感知形式。对应于前述环境
种类的数目，有多个的输入参数。其中一个参数是从发射机到接收机的归一化距离。其
余的参数则是基于连接收发信机的直线上所涉及的环境种类的分，即多少个门，这条线
上多大比例是通过教室等等。该模型的输出是归一化的场强电平。
Determining the parameters of the ANN model is very simple. Statistical anal
ysis is unnecessary. The neural network should just be trained with the meas
ured data. Computationally the training process is very extensive, but it is
done only once. In implementation, it was shown that the accuracy of the AN
N model is comparable to the accuracy of the ray-tracing and FDTD models.
确定ANN模型的参数非常简单。不必要进行统计分析。可以用测试数据对神经网络进行训
练。训练过程中的计算量可能会很大，但这只需要做一次。在实现中，有结果表明ANN模
型的精确度可与射线追踪及FDTD模型的精度相比拟。
Additional Effects
其他效果
Although previously described models do not take into account temporal fadin
g and the effects of the human body, these phenomena must be carefully consi
dered in a proper design of an indoor radio communication system.
尽管前述模型没有考虑时域衰落及对人体的影响，在进行室内无线通信系统设计中必须
对这些现象加以谨慎考虑。
Extensive measurements carried out at the School of Electrical Engineering i
n Belgrade [38] have shown that a portable terminal slowly moving through an
indoor environment experiences Ricean or Rayleigh fading depending on wheth
er LoS conditions exist or not. In the former case, Ricean K factor takes va
lues in the range 2 to 10dB. Almost identical results are obtained by Alexan
der [42] and Rappaport and McGillem [43].
在贝尔格莱德大学电子工程学院进行的广泛测量表明[38]，根据直达射线存在与否的不
同，手持终端在室内环境缓慢移动时会经历Ricean或Rayleigh衰落。在前一情形下，Ri
cean K因子的取值范围为2到10dB。Alexander [42]、Rappaport及McGillem[43]几乎得
到了相同的结果。
The influence of a human body on signal reception can be viewed through two
main aspects [46]. The first is the influence of the user's own body and the
second is the effect of bodies of the people who are in the vicinity of the
user. On average, both effects cause a drop in the received signal level, w
hile the latter, in addition, causes the received signal level to fluctuate
more irregularly. These irregularities are very hard to predict due to their
great dependence on the exact arrangement of the people near the user.
人体对信号接收的影响可以从两个主要方面来看[46]。包括户自身身体的影响以及用户
附近其他人身体的影响。平均来看两种情况都会造成接收信号的下降，而后者则使接收
信号电平的波动更不规则。由于这种不规则变化对用户附近人体的准确分布简有极大的
依赖性，因此极难加以预测。
For fixed transmitter and receiver positions, the ambient motion causes temp
oral fading, which can be described as Ricean fading, with the Ricean K fact
or of about 10dB [40]. In this case, typical variations of the signal level
are less than 15dB for 99.9 percent of the time, which is slightly greater t
han the variations typically obtained in an outdoor environment [9].
当收发信机固定时，周围运动会导致时域衰落，这可以归为Ricean衰落且Ricean K因子
约10dB [40]。这一情形下，信号电平的典型变化在99.9%的时间内小于15dB，这略大于
室外环境下所得到的典型变化值。
Conclusion
结论
Today's fast growing radio communication market places stronger and stronger
demands on the design of a radio system. A proper system design requires ac
curate and reliable radio channel models, among which the field strength pre
diction models are most important. Although many researchers have been worki
ng hard during the past few decades in the area of field strength prediction
, there are still numerous problems to be solved. To date, a perfect propaga
tion model has not been found. Instead, many different prediction tools have
been proposed. Each of them has advantages and disadvantages and can be app
lied only to particular circumstances. Further investigations of methods for
increasing accuracy and decreasing computational time are necessary. Howeve
r, it should be noted that the progress in field strength prediction models
depends to a great extent on the development of environmental databases as w
ell as on the development of computational resources.
当今快速发展的无线通信市场对无线系统的设计提出了越来越多的要求。适当的系统设
计需要准确可靠的无线信道模型，其中，场强预测模型又最为重要。尽管很多研究者在
过去的数十年间在场强预测领域辛勤工作，但仍有很多问题尚未解决。到目前为止，还
没有发现一种完美的传播模型。事实上是提出了很多不同的预测工具。这些工具都有各
自的优缺点且智能应用于特定的环境。在提高精确度和降低计算时间方面需要有对有关
方法的进一步研究。应当注意到，场强预测模型的进展在很大程度上依赖于环境数据库
的发展以及可供适用的计算资源的发展。
References
[1] W. C. Y. Lee and Y. S. Yeh, "On the Estimation of the Second-Order Stati
stics of Log Normal Fading in Mobile Radio Environment," IEEE Trans. Commun.
, June 1974, pp. 869-73.
[2] K. Bullington, "Radio Propagation for Vehicular Communications," IEEE Tr
ans.Vehic. Tech., vol. VT-26, no. 4, Nov. 1977, pp. 295-308.
[3] Y. Okumura et al., "Field Strength and its Variability in VHF and UHF La
nd-Mobile Services," Review Elec. Commun. Labs., vol. 16, Sept.-Oct., 1968,
pp. 825-37.
[4] CCIR, XVIIth Plenary Assembly, Duselldorf, 1990, vol. V, Rep. 567-3, Rec
. 370-4, Rep. 239-4.
[5] M. Hata, "Empirical Formula for Propagation Loss in Land Mobile Radio Se
rvices," IEEE Trans. Vehic. Tech., vol. 29, no. 3, 1980.
[6] CCIR Study Groups Period 1978-1982, European Broadcasting Union, "Improv
ement of Predictions of Field Strengths in VHF and UHF Bands," Doc. 5/28-E,
1980.
[7] CCIR Study Groups Period 1978-1982, Peoples Republic of Poland, "VHF and
UHF Propagation Curves in the Frequency Range 30MHz to 1GHz for the Broadca
sting Services," Doc. 5/79-E, 1980.
[8] A. G. Longley and P. L. Rice, "rediction of Tropospheric Radio Transmis
sion Loss Over Irregular Terrain -- A Computer Method -- 1968," Inst. Teleco
mmun. Sci., ESSA Tech. Rep ERL79-ITS67, Boulder, CO.
[9] W. C. Y. Lee, Mobile Communications Design Fundamentals, John Wiley, 199
3.
[10] IEEE Vehic. Tec. Society, Special Issue on Mobile Radio Prop., IEEE Tra
ns. Vehic. Tech., vol. 37, 1988, pp. 3-72.
[11] Commission of the European Communities, "Digital Land Mobile Radio Comm
unications COST 207," 1984.
[12] M. Ibrahim and J. Parsons, "Signal Strength Prediction in Built-up Area
s," IEE Proc., vol. 130, pt. F, no. 5, 1983.
[13] M. Atefi and J. Parsons, "Urban Radio Propagation in Mobile Radio Frequ
ency Bands," IEE Commun. 86, Conf. Pub. no. 262, 1986.
[14] C. Dadson, J. Durkin, and R. Martin, "Computer Prediction of Field Stre
ngth in the Planning of Radio Systems," IEEE Trans. Vehic. Tech., vol. 24, n
o. 1, 1975.
[15] J. Walfisch and H. L. Bertoni, "A Theoretical Model of UHF Propagation
in Urban Environments," IEEE Trans. Ant. Prop., vol. 36, Dec. 1988, pp. 1788
-96.
[16] F. Ikegami et. al., "Theoretical Prediction of Mean Field Strength on U
rban Streets," IEEE Trans. Ant. and Prop., vol. 32, no. 8, 1984.
[17] F. Ikegami et al., "ropagation Factors Controlling Mean Field Strength
on Urban Streets," IEEE Trans. Ant. and Prop., vol. 32, no. 8, 1984.
[18] R. Grosskopf, "rediction of Urban Propagation Loss," IEEE Trans. Anten
nas and Prop., vol. 42, no. 5, 1994.
[19] A. Neskovic, N. Neskovic, and D. Paunovic, "A Field Strength Prediction
Model Based on Artificial Neural Networks," Proc. 9th IEEE Med. Electro. Co
nf. - MELECON, Tel Aviv - Israel, May 1998.
[20] "EET-Ericsson Engineering Tool, User Reference Guide," Ericsson Radio S
ystems AB, 1997.
[21] K. Allsebrook and J. D. Parsons, "Mobile Radio Propagation in British C
ities at Frequencies in the VHF and UHF Bands," IEEE Trans. Vehic. Tech., vo
l. VT-26, no. 4, Nov. 1977, pp. 313-23.
[22] A. Mehrotra, Cellular Radio Performance Engineering, Artech House, 1994
.
[23] D. Paunovic, Z. Stojanovic, and I. Stojanovic, "Choice of a Suitable Me
thod for the Prediction of the Field Strength in Planning Land Mobile Radio
Systems," IEEE Trans. Vehic. Tech., vol. 33, no. 3, 1984, pp. 259-65.
[24] A. Neskovic, N. Neskovic, and D. Paunovic, "Measurements and Analysis o
f the Electric Field Level in Band 450MHz," Project of Power Distribution Co
mpany - Belgrade, 1996.
[25] M. Vujovic et al., "Determining Parameters of the Lee Prediction Model,
" Proc. 43th Annual Conf. ETRAN, Zlatibor--Yugoslavia, June 1997.
[26] "ropagation Prediction Models," COST 231 Final Rep., ch.4, pp. 17-21.
[27] D. Har, A. M. Watson, and A. G. Chadney, "Comment on Diffraction Loss o
f Rooftop-to-Street in COST 231 -- Walfisch-Ikegami Model," IEEE Trans. Vehi
c. Tech., vol. 48, no. 5, Sep. 1999, pp. 1451-52.
[28] O. Perrault , J. P. Rossi, and T. Balandier, "redicting Field Strength
With a Neural Ray-Tracing Model," Proc. IEEE GLOBECOM, London, Nov. 1996.
[29] H. H. Mohamad, "Fundamentals of Artificial Neural Networks," Massachuse
tts Institute of Technology, 1995.
[30] P. Harley, "Short Distance Attenuation Measurements at 900MHz and 1.8GH
z Using Low Antenna Heights for Microcells," IEEE JSAC, vol. 7, no. 1, Jan.
1989.
[31] F. Lotse and A. Wejke, "Propagation Measurements for Microcells in Cent
ral Stockholm," IEEE Vehic. Tech. Conf., 1990.
[32] H. H. Xia et al., "Radio Propagation Characteristics for Line-of-Sight
Microcellular and Personal Communication," IEEE Trans. Ant. and Prop., vol.
41, no. 10, Oct. 1993.
[33] R. J. Luebbers, "Finite Conductivity Uniform GTD Versus Knife-edge Diff
raction in Prediction of Propagation Path Loss," IEEE Trans. Ant. Prop., vol
. AP-32, Jan. 1984, pp. 70-72.
[34] R. G. Kouyoumijan and P. H. Pathak, "A Uniform Theory of Diffraction fo
r an Edge in a Perfectly Conducting Surface," Proc. IEEE, vol. 62, Nov. 1974
, pp. 1448-61.
[35] S. Y. Tan and H. S. Tan, "A Microcellular Communications Propagation Mo
del Based on the Uniform Theory of Diffraction and Multiple Image Theory," I
EEE Trans. Ant. and Prop., vol. 44, no. 10, Oct. 1996.
[36] N. Neskovic, "The Field Strength Prediction for Urban Microcells (band
900MHz)," Master's Thesis, Faculty of Electrical Engineering -- Belgrade, 19
97.
[37] N. Neskovic, A. Neskovic, and D. Paunovic, "A New Microcell Prediction
Model Based on the Arrangement of the Streets and their Types," Proc. 10th I
EEE Med. Electrotech. Conf.- MELECON, Limassol - Cyprus, May 2000.
[38] A. Neskovic, "Artificial Neural Network Indoor Radio Propagation Model
(band 900MHz)," Master's Thesis, Faculty of Electrical Engineering -- Belgra
de, 1997.
[39] P. Constatinou, "Properties of Wireless RF Channels," Wireless LAN Syst
ems, Artech House, pp. 129-63, 1994.
[40] A. B. Jorgen, T. S. Rappaport, and S. Yoshida, "Propagation Measurement
s and Models for Wireless Communications Channels," IEEE Commun. Mag., vol.
33, no. 1, Jan. 1995, pp. 42-49.
[41] K. Pahlavan and A. H. Levesque, Wireless Information Networks, New York
: John Wiley & Sons, Inc., 1995.
[42] S. E. Alexander, "900MHz Propagation Within Buildings," IEE Second Int'
l Conf. Radio Spectrum Conservation Techniques, 1983, pp. 51-55.
[43] T. S. Rappaport and C D. McGillem, "UHF Fading in Factories," IEEE JSAC
, vol. 7, no. 1, Jan. 1989, pp. 40-48.
[44] A. A. M. Saleh and R. A. Valenzuela, "A Statistical Model for Indoor Mu
ltipath Propagation," IEEE JSAC, vol. 5, no. 2, Feb. 1989, pp. 128-137.
[45] A. Neskovic, N. Neskovic, and D. Paunovic, "Indoor Electric Field Level
Prediction Model Based on the Artificial Neural Networks," IEEE COMML, vol.
4, no. 6, June 2000.
[46] A. Neskovic, N. Neskovic, and D. Paunovic, "Influence of the Human Body
on Indoor Radio Communications at 450MHz -- Measurements and Analysis," Pro
c. 43th Annual Conf. ETRAN, Budva -- Yugoslavia, June 1996.
Biographies
Aleksandar Neskovic received his Dipl.Eng. and M.S.E.E. degrees from the Sch
ool of Electrical Engineering, University of Belgrade, Yugoslavia, in 1993 a
nd 1997, respectively. Since Sept. 1993 he has been a research and teaching
assistant with the Dept. of Communications of the School of Electrical Engin
eering, Belgrade, Yugoslavia. His research is focused on the radio-system de
sign, investigation and modeling of mobile radio propagation and development
of simulation methods for a mobile radio channel. He has published several
papers in proceedings of international conferences. During the last five yea
rs he has been fully involved in several projects including the design of pu
blic (GSM) and private (TETRA) mobile radio systems as well as in designing
radio and TV broadcasting systems. These projects have been mainly conducted
by major national telecommunication and power supply companies.
Natasa Neskovic received her Dipl.Eng. and M.S.E.E. degrees from the School
of Electrical Engineering, University of Belgrade, Yugoslavia, in 1993 and 1
997, respectively. Since Sept. 1993 she has been a research and teaching ass
istant with the Department of Communications of the School of Electrical Eng
ineering, Belgrade, Yugoslavia. Her research is focused on radio-system desi
gn, the investigation and modeling of

---

时间:  2004-12-18 14:27

作者: 固定无线接入

不错啊，可以联系阅读

---

时间:  2004-12-18 16:06

作者: mick

太多了，没有时间呀

---

通信人家园 (https://www.txrjy.com/)

Powered by C114