81
Chapter 5
Modeling of theWireless
Channels with Leaky
Waveguide for
Communications-Based
Train Control Systems
Hongwei Wang, F.Richard Yu, Li Zhu, and Tao Tang
Contents
5.1 Introduction ...............................................................................................82
5.2 Leaky Waveguide in CBTC Systems ..........................................................82
5.2.1 Overview of CBTC Radio Channel with Leaky Waveguide ...........82
5.3 Measurement Campaign ............................................................................83
5.4 Modeling of the CBTC Radio Channel with Leaky Waveguide ............... 84
5.4.1 Measurement Results ..................................................................... 84
5.4.2 Determination of the Path Loss Exponent ......................................85
5.4.3 Determination of the Small-Scale Fading .......................................88
5.5 Conclusion .................................................................................................91
References ...........................................................................................................91
82 ◾ Advances in Communications-Based Train Control Systems
5.1 Introduction
As urban rail transit systems are deployed in a variety of environments (subway tunnels,
viaducts, etc.), there are dierent wireless network congurations and propagation
media. For the viaduct scenarios, leaky rectangular waveguide is a popular approach,
as it can provide better performance and stronger anti-interference ability than the free
space [1]. For example, leaky waveguide has been applied in Beijing Subway Yizhuang
Line. In addition, due to the available commercial o-the-shelf equipment, wireless
local area networks (WLANs) are often adopted as the main method of train–ground
communications for communications-based train control (CBTC) systems [2].
For general applications, leaky waveguide is taken as a leaky wave antenna with
the length of several operation wavelengths. However, in CBTC systems, the length
of leaky waveguide is several hundred meters and the distance between the leaky
waveguide and the receiving antenna is short (about 30cm). Due to the specicity
of the CBTC application, there are only a few research works about leaky wave-
guide in CBTC systems. e author of [1] gives a description of leaky waveguide
used in CBTC systems, and the advantages of leaky waveguide are demonstrated
by comparisons with natural propagation. e characteristics of leaky waveguide
are shown in [3] through laboratory measurements.
In this chapter, based on the measurement results in Beijing Subway Yizhuang Line,
we use the polynomial tting and equivalent magnetic dipole method to build the path
loss model. In addition, the Akaike information criterion with a correction (AICc) is
applied to determine the distribution model of the small-scale fading. e proposed
path loss model of the channel with leaky waveguide in CBTC systems is linear, and the
path loss exponent can be approximated by the transmission loss of leaky waveguide.
We show that the small-scale fading follows log-normal distribution, which is often
referred to as the distribution model of the small-scale fading in body area communi-
cation propagation channels [4,5]. In addition, the corresponding parameters of log-
normal distribution μ
dB
and
σ
dB
are also determined from the measurement results.
e rest of this chapter is organized as follows: Section 5.2 describes an over-
view of CBTC systems and the application scenario of leaky waveguide in CBTC
systems. Section 5.3 discusses the real eld measurement conguration and sce-
nario. en, Section 5.4 presents the path loss model and the small-scale fading
model. Finally, Section 5.5 concludes the chapter.
5.2 Leaky Waveguide in CBTC Systems
5.2.1 Overview of CBTC Radio Channel
with Leaky Waveguide
Generally speaking, the radio waves of CBTC systems are often transmitted in
the free space, especially in tunnels. However, in the viaduct scenarios, the per-
formance of wireless communication could be aected by the interference from
Modeling of the Wireless Channels with Leaky Waveguide ◾ 83
other wireless devices in surrounding buildings. ere are periodic transverse slots
in the wide wall of leaky waveguide, which can provide stable signals and anti-
interference ability. As a result, leaky waveguide has gradually been applied as the
propagation medium in CBTC systems. It can be deployed along the track, as
shown in Figure5.1. Considering the unique characteristics of leaky waveguide in
CBTC systems, we propose a model of the channel with leaky waveguide to facili-
tate the design and evaluation of the performance of CBTC systems in this chapter.
5.3 Measurement Campaign
Our measurements were performed at the section from Rongjingdong Station to
Tongjinan Station of Beijing Subway Yizhuang Line under real operation conditions.
Two sets of Cisco WLAN devices are used, one of which is set as the access point (AP)
and the other one is set as the mobile station (MS). Both of them are set to work at the
frequency of 2.412GHz, which is also called channel 1. e output power of the AP
taken as the transmitter is set as 30dBm. rough a coupling unit, the AP is connected
with the leaky waveguide. e MS taken as the receiver is located on a measurement
vehicle with a panel antenna to receive the signals leaked from the leaky waveguide. e
gain of the antenna is 11dBi and the beam width is 30°. Due to the limits of subway
line and the restricted conditions of trains, especially the Bogie, the distance between
the antenna and the leaky waveguide is within the scope of 300–500mm. In our mea-
surements, the receiving antenna was xed on the measurement vehicle 320mm above
the leaky waveguide in order to capture the channel samples in the near eld of the
receiving antenna. e length of one section of leaky waveguide is about 300m.
e location of the receiver is obtained through a velocity sensor installed on
the wheel of the measurement vehicle, which can detect the real-time velocity, and
Figure 5.1 Leaky waveguide applied in viaduct scenarios of Beijing Subway
Yizhuang Line.
84 ◾ Advances in Communications-Based Train Control Systems
the resolution of position is millimeter per second. e measurement vehicle moved
along the tracks with the velocity of 70km/h. Figure 5.2 shows the measurement
equipment used in our measurements. Figure 5.3 shows the measurement scenario.
5.4 Modeling of the CBTC Radio Channel
with Leaky Waveguide
5.4.1 Measurement Results
We have performed 20 measurements to obtain the statistical characteristics of
the channel with leaky waveguide. We use the polynomial ttings of degree 1 and
degree 2 to process the experimental data. In the data processing, we nd that the
quadratic coecients of polynomial ttings of degree 2 are very small (about 10
–4
),
Panel antenna
Leaky waveguide
Figure5.2 Measurement equipment used in the CBTC channel measurements.
Leaky waveguide
Feeder line
Coupling unit
Measurement vehicle
Single chip
Velocity
Velocity sensor
MS
AP
Ethernet
(signal strength and SNR)
Serial port
(displacement)
On-board panel antenna
Figure5.3 Measurement scenario.
Modeling of the Wireless Channels with Leaky Waveguide ◾ 85
which means that the quadratic coecients of the polynomials can be ignored.
Hence, the fading channel with the leaky waveguide can be approximated as a
linear model according to the polynomial tting results.
Some classic channel models have been proposed for dierent environments, such
as the Okumura–Hata model, the COST 231 model, and the Motley–Keenan model.
Due to the specicity of the CBTC application, to the best of our know ledge, there is
a channel model with leaky waveguide in CBTC systems. As the channel with leaky
waveguide is linear as shown earlier, similar to the expression of the Okumura–Hata
model [6], we propose a model of the CBTC channel with leaky waveguide as follows:
PL PL(0)
ss
()
dndX=++
(5.1)
where:
PL(0) is the dierence between the power leaked from the beginning of the
leaky waveguide and the input power
n is the path loss exponent whose unit is dB/m
d is the location of the receiving antenna relative to the beginning of the leaky
waveguide
X
ss
is the small-scale fading the is a random variable
PL(0) depends on the dimensions of the leaky waveguide and the slots. ere are
two key parameters, n and X
ss
, in the model that will be determined in Section 5.4.2
and 5.4.3 respectively.
5.4.2 Determination of the Path Loss Exponent
According to the slopes of tting lines, we can get the average value of the path loss
exponent as follows:
n
t
n
m
t
m
=
∑
1
(5.2)
where:
t
is the total number of measurements
n
m
is the slope of the tting line of the m th measurement
In our measurements, the average value of the path loss exponent is
0.0136
/dB
m
.
e consecutive slot radiation can be represented by separate successive equiva-
lent magnetic dipoles fed by the same power [1]. e equivalent method can be eec-
tive to describe the fading tendency of the channel with leaky waveguide, but it may
not be reasonable to assume that the magnetic dipoles are fed by the same power
when the length of the leaky waveguide is so large, because the transmission loss
should not be ignored. en, we need to calculate the transmission loss as follows [1]:
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
18.188.200.46