没有合适的资源?快使用搜索试试~ 我知道了~
A Statistical Theory of Mobile-Radio reception.pdf
5星 · 超过95%的资源 需积分: 10 13 下载量 76 浏览量
2015-12-24
19:09:16
上传
评论
收藏 9.72MB PDF 举报
温馨提示
试读
44页
A Statistical Theory of Mobile-Radio reception.pdf
资源推荐
资源详情
资源评论
A Statistical Theory of Mobile-Radio
The statistical characteristics of the fields and signals in the reception
of radio frequencies by a moving vehicle are deduced from a scattering propa-
gation model. The model assumes that the field incident on the receiver
antenna is composed of randomly phased azimuthal plane waves of arbi-
trary azimuth angles. Amplitude and phase distributions and spatial
correlations of fields and signals are deduced, and a simple direct rela-
tionship is established between the signal amplitude spectrum and the
product of the incident plane waves' angular distribution and the azimuthal
antenna gain.
The coherence of two mobile-radio signals of different frequencies is
shown to depend on the statistical distribution of the relative time delays
in the arrival of the component waves, and the coherent bandwidth is shown
to be the inverse of the spread in time delays.
Wherever possible theoretical predictions are compared with the experi-
mental results. There is sufficient agreement to indicate the validity of the
approach. Agreement improves if allowance is made for the nonstationary
character of mobile-radio signals.
I. INTRODUCTION
In a typical mobile-radio situation one station is fixed in position
while the other is moving, usually in such a way that the direct line
between transmitter and receiver is obstructed by buildings. At ultra-
high frequencies and above, therefore, the mode of propagation of the
electromagnetic energy from transmitter to receiver will be largely
by way of scattering, either by reflection from the flat sides of build-
ings or by diffraction around such buildings or other man-made o r
natural obstacles.
i.i The Model
It therefore seems reasonable to suppose that at any point the
received field is made up of a number of generally horizontally trav-
By R. H. CLARKE
957
958 THE BELL SYSTEM TECHNICAL JOURNAL, JULY-AUGUST 196 8
eling free-space plane waves whose azimuthal angles of arrival occur
at random for different positions of the receiver, and whose phases
are completely random such that the phase is rectangularly distributed
throughout 0 to 2w. The phase and angle of arrival of each component
wave will be assumed to be statistically independent. The probability
density function ρ (a) which gives the probability ρ {a) da that a com-
ponent plane wave will occur in the azimuthal sector from a to a + da
will not be specified, since it will b e different for different environ-
ments, and is also likely to vary from region to region within one
environment; but the assumption that the phase φ has a rectangular
probability density function throughout 0 to 2* will be made in all
cases.
For simplicity, it will be assumed that at every point there are
exactly Ν component waves and that these Ν waves have the same
amplitude. In addition it will be assumed that the transmitted radia-
tion is vertically polarized, that is, with the electric-field vector di-
rected vertically, and that the polarization is unchanged on scattering
so that the received field is also vertically polarized.
The model described so far gives what might be termed the "scat -
tered field," since the energy arrives at the receiver by way of a
number of indirect paths. Another term for this scattered field is the
"incoherent field," because its phase is completely random. Some-
times a significant fraction of the total received energy arrives by
way of the direct line-of-sight path from transmitter to receiver . The
phase of the "direct wave" is nonrandom and it may therefore be
described as a "coherent wave." It will be seen later that the field
in a heavily built-up area such as New York City is entirely of the
scattered type, whereas the field in a suburban area with the trans-
mitter not more than a mile or two distant is often a combination o f
a scattered field with a direct wave.
1.2 Comparison With Other Proposed Models
J. F. Ossanna
1
was the first to attempt an explanation of the sta-
tistical character of the received mobile-radio signal in terms of a set
of interfering waves. He was concerned with measurements taken in
a suburban environment, and assumed that reflection occurred at the
flat sides of houses and that the incident and reflected waves form an
interference pattern through which the receiver moves. He the n as-
sumed that all orientations of the sides of houses are equally likely,
and hence obtained spectra for the randomly fading signal wit h the
MOBILE RADIO
959
angle between the direction of vehicle motion and the direction to
the transmitter as a parameter.
There is quite good agreement between Ossanna's theoretical spectra
and those derived from measurements on several suburban streets
situated within 2 miles of the transmitter. There i s marked disagree-
ment, however, at very low frequencies and at frequencies i n the
region of the sharp cut-off associated with the maximum Doppler
frequency shift. At very low frequencies the spectral energy is al-
ways observed to be higher than that predicted by theory, whether
Ossanna's or the one we use in this paper. The reason for this is that
neither theoretical model takes into account the large-scale varia-
tions in total energy which result from the changing topography
between transmitter and mobile receiver.
The basic difference between Ossanna's theoretical model and the
model used here is that the former is essentially a reflection model
whereas the latter is essentially a scattering model and so include s
the former as a special case. An example of the limitations of the
reflection model can be seen from the experimental spectra plotted
in Ossanna's paper. The spectra are derived from signal-fading rec-
ords made on several streets whose inclination to the transmitter
direction ranged from 15 degrees to 84 degrees, and in each case
there is evidence of a shelf which cuts off at twice th e maximum
Doppler frequency shift. Ignoring the higher harmonics generated in
the detection process, the reflection model predicts a spectral cutoff
which depends on the direction of the street with respect to the trans-
mitter, ranging from the maximum Doppler frequency shift itself
when the street is at right angles to the transmitter direction to twice
that value when the street is in line with the transmitter.
With the scattering model, on the other hand, the angular distribu-
tion ρ (a) of scattered waves can be chosen to predict the existence
of a spectral shelf out to twice the Doppler frequency shift for any
street direction. Another feature of the reflection model which makes
it rather inflexible is that for every randomly oriented reflected wave
there exists a direct wave incident on the mobile receiver and carry-
ing the same power. Thus the ratio of coherent to incoherent power
in the received signal is fixed, whereas in the scattering model this
ratio is arbitrary and may be adjusted according to the environment.
In his study of energy reception in mobile radio, Ε. N. Gilbert*
examined several models of the scattering typ e and established a
number of important relationships between them. One feature com-
960 THE BELL SYSTEM TECHNICAL JOURNAL, JULY-AUGUST 1968
mon to all of them, however, was the uniform distribution of waves
in angle, although he briefly mentioned the effect of a single strong
component arriving directly from the transmitter. The first model
Gilbert considered was that of Ν waves arriving from fixed directions,
equally spaced in angle. The phases of the waves were assumed to be
independent and uniformly distributed throughout 0 to 3ir; their
amplitudes were assumed to be Rayleigh distributed and independent,
but with the same variance. In a second model the angles o f arrival
were allowed to occur at random with equal probability for any
direction; the phases were again completely random but the ampli-
tudes were assumed to be constant. (This model is the same as the
one we use in this paper, with the restriction that p[a] = [2π·]
-1
.) A
third model was an extension of the second to include the case of an
arbitrary distribution of the amplitudes. Gilbert showed that the
second and third models were equivalent to the first for sufficiently
large N.
1.3 Scope
This paper shows that the scattering model ca n be used to predict
the statistical characteristics of the signal received at the antenna
terminals, hence at the output of a square-law or envelope detector,
of the mobile receiving vehicle. These characteristics include the
probability distributions of amplitude and phase, spatial correlations,
amplitude spectra, and frequency correlations.
A simple relationship is established betwee n the spectrum of the
signal input and the product of the azimuthal power gain g[a) of the
antenna and the probability distribution function ρ (a) of the angle
of arrival of the component waves. This relationship will be particu-
larly useful in analyzing mobile-radio systems with directional an-
tennas on the mobile unit.
Other topics discussed are the use of space and frequency diversity,
coherent bandwidth, and random frequency modulation. Some com-
ments also are made on the nonstationary aspects of mobile-radio
fields and on the consequent need for their characterization in terms
which will be useful to the mobile-radio system designer. Whenever
possible the theory is discussed in the light of available experiments.
II.
FIRST-ORDER STATISTICS OF THE FIELD
2.1
Theory
Under the assumption that the total field at any receiving point
is vertically polarized and is composed o f the superposition of Ν
MOBILE RADIO
961
waves, the n
tb
wave arriving at any angle a„ to the χ axis (Fig. 1)
with phase φ
α
, the field components at point 0 (the zero phase refer-
ence point) are
Ε,
= Ε„Σ exp \kn\ 0)
n-l
H
- = - 7
E
sin
«»exp
(2 )
Ε
y
Η, = — Σ cosa, exp (jV.|. (3 )
In these equations E
0
is the common (real) amplitude of the
Λ Γ
waves
and η is the intrinsic impedance of free space. The time variation is
understood to be of the form exp{j<at). Notice that E, will be propor-
tional to the signal input to the receiver when a vertical dipole an-
tenna is used, and that E
z
, H
T
, and H
v
will be proportional to the
three inputs from a Pierce antenna system.
2
The three field components E
z
, H
x
, and H
v
are complex Gaussian
random variables, to a good approximation, provided tha t Ν is
suf-
ficiently large. This is a consequence of the Central Limit Theorem
and the assumption that the phases φ„ are independent of each other
and of the angles of arrival α». Thus each field component has a real
part and an imaginary part whic h are approximately zero-mean Gaus-
sian random variables of equal variance, the approximation improv-
ing for larger N, and provided that the phases
<p
n
are rectangularly
distributed throughout 0 to 2v. Appendix A shows that under the
same assumptions the real and imaginary parts of each field com-
ponent are uncorrelated; the y are therefore approximately statistically
independent.
8
An important consequence of this is that the envelope of all three
field components (hence of the signals at the terminals of a vertical
Fig. 1 — A typical component wave an d the two field points 0 and 0* .
剩余43页未读,继续阅读
资源评论
- qq_387651532019-03-16难得的资源,谢谢分享!
csc270
- 粉丝: 0
- 资源: 1
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功