HANOI UNIVERSITY OF SCIENCE AND TECHNOLOGY

SCHOOL OF ELECTRICAL AND ELECTRIC ENGINEERING

FINAL EXAM
5G channel modeling

Student : Ngo Vu Truong Son

Student ID : 20203560
Class : ET1 – 01 – K65
Lecture : Ph.D Nguyen Thu Nga

Hanoi, 2-2023
TABLE OF CONTENTS

ABBEVIATION LIST…………………………………………………2
SYMBOLS LIST……………………………………………………….3
FIGURES LIST………………………………………………………...4
1. COORDINATE SYSTEM………………………………………….5
1.1.
Definition……………………………………………………………………………..5
1.2. Scenarios of
interest………………………………………………………………….6

2. THE SIMULATION OF CORRELATION PROPERTIES OF

ANTENNA IN BS SIDE AND MS SIDE IN CASE OF NLOS……..7
2.1. The correlation functions of the 5G channel modelling in case of Non-Line of
sight (NLOS)……………………………………………………………………...
…………….7
2.2. The correlation properties in BS
side……………………………………………….8
2.3. The correlation properties in MS side………………………………………………
9

3. CONCLUSION…………………………………………………….10
4. REFERENCES……………………………………………………..10

1
ABBEVIATION LIST

3D Three-dimensional
AOA Azimuth angle Of Arrival
AOD Azimuth angle Of Departure
BS Base station
GCS Global Coordinate System
LCS Local Coordinate System
MIMO Multiple In, Multiple Out
MS Mobile station
NLOS Non-line of sight
O2O Outdoor-to-Indoor
O2I Outdoor-to-Indoor
RMa Rural Macro
UMi Urban Micro
ZOA Zenith angle Of Arrival
ZOD Zenith angle Of Departure

2
SYMBOLS LIST

f frequency
n^ sphercial unit vector
α bearing angle
β down-tilt angle
γ slant angle
ϕ azimuth angle
θ zenith angle
ϕ^ spherical basis vector (unit vector) for GCS
ϕ^ spherical basis vector (unit vector) for LCS
'

θ^ spherical basis vector (unit vector), orthogonal to ϕ^ , for GCS

θ^
'
spherical basis vector (unit vector), orthogonal to ϕ^' , for LCS
ψ angular displacement between two pairs of unit vectors
Σ multiple sum
ρ correlation function
Δ ds antenna element’s spacing in BS
Δ du antenna element’s spacing in MS
Δ𝑡 time difference between two observations
Δf frequency difference
F rx ,u ,θ radiation field of the receive antenna element 𝑢 with direction θ
F rx ,u ,ϕ radiation field of the receive antenna element 𝑢 with direction ϕ
F tx , s ,θ the radiation field of transmit antenna element 𝑠 with direction θ
F tx , s ,ϕ the radiation field of transmit antenna element 𝑠 with direction ϕ
T
r^ tx ,n , m the spherical unit vector as the angle ϕ n , m , AOA and θn , m , ZOA
T
r^ tx ,n , m the spherical unit vector as the angle ϕ n , m , AOD and θn , m , ZOD
d rx ,u the location vector of antenna element u
d tx ,s the location vector of antenna element s

FIGURES LIST

3
Figure 1. The Cartesian of 5G channel modeling [1]
Figure 2. The correlation properties in BS side [2]
Figure 3. The correlation properties in MS side [3]

1. COORDINATE SYSTEM
1.1. Definition

4
 Figure 2. The Cartesian of 5G channel modeling [1]

The diagram shown in Figure 1 depicts x, y, and z axes, spherical angles, and
spherical unit vectors in a coordinate system. Additionally, it shows that the zenith angle,
denoted by θ, and azimuth angle, denoted by φ, are defined. The use of multiple Base
Stations (BSs) and User Terminals (UTs) constitutes what is known as a Global
Coordinate System (GCS). On the other hand, a Local Coordinate System (LCS) serves
as a reference frame for defining the far-field vector characteristics, such as the pattern
and polarization, of each antenna element in an array antenna utilized by either a BS or
UT.
The orientation of an array relative to the Global Coordinate System (GCS) is
characterized by a series of rotations. Transforming the vector properties of the array
elements from the Local Coordinate System (LCS) to the GCS is crucial, as this is solely
determined by the array's orientation, which often diverges from the orientation of the
GCS. It's worth mentioning that by rotating the LCS in relation to the GCS, any desired
mechanical alignment of the array can be achieved.
We have posited the existence of a GCS with coordinates ( x , y , z , θ , φ ) and unit
vectors (θ^ ,ϕ^ ), as well as a LCS with "primed" coordinates ( x ' , y ' , z ' ,θ ' , ϕ ′) and unit
vectors ( θ^' , ϕ^' ). The arbitrary 3D rotation from the LCS to the GCS is described by three
angles: α, which defines the bearing, β, which defines the down-tilt, and γ, which defines
the slant.

5
 After the first rotation of α about the z-axis, the ẏ axis is returned to its original
state, and after the first rotation about z and the second rotation of β about ẏ , the ẍ axis is
reset to its initial position. The antenna bearing angle is established after the first rotation
about z. Following the second rotation of β about ẏ , the antenna is tilted downward. The
third rotation of γ about ẍ determines the antenna slant angle. The orientation of the x, y,
and z axes after all three rotations is depicted using the symbols x, y. The angle necessary
for rotation to the GCS is provided as described in [1].
ψ=arg ( sinγcosθ sin ( ϕ−α ) +cosγ ( cosβsinθ−sinβcosθ cos ( ϕ−α ) ) + j ( sinγ cos ( ϕ−α ) +sinβcosγ sin ( ϕ−α ) ) )
(1)
1.2. Scenarios of interest
Based on [1], which provides a detailed description of three scenarios used, namely
Urban Micro (UMi), Indoor, and Rural Macro (RMa), ...
The UMi scenario includes Street canyon and Open area, with O2O (Outdoor-to-
Outdoor) and O2I (Outdoor-to-Indoor). In this scenario, the base stations (BSs) are
installed below the rooftop levels of nearby buildings. UMi Open area is specifically
designed to replicate real-life scenarios for distances of 50 to 100 meters.
The Indoor scenario aims to simulate different indoor deployment scenarios such as
office spaces and shopping malls. In this scenario, the base stations (BSs) are typically
mounted on ceilings or walls at a height of 2-3 meters. Shopping malls, which are usually
several stories high and have multiple floors, are also included in this scenario.
The RMa scenario is dedicated to rural deployments, with a focus on providing
extensive and continuous coverage for high-speed vehicles under noise-limited and/or
interference-limited conditions. This is achieved through the use of macro transmission
reception points.

6
 2. THE SIMULATION OF CORRELATION
PROPERTIES OF ANTENNA IN BS SIDE AND MS
SIDE IN CASE OF NLOS
2.1. The correlation functions of the 5G channel modelling in
case of Non-Line of sight (NLOS)
The antenna system is a 2 × 2 configuration, comprising of two transmit antennas
on the BS and two receive antennas on the MS. The channel transfer function H u , s ( f , t )
is derived from a Fourier transform of hu , s ( τ , t ) , which is defined for s antenna
elements on the BS side and u antenna elements on the MS side. This is in accordance
with the axes system outlined in section 1.1 and is described as follows in reference :
2 3 N

, s ( τ ,t ) = ∑ ∑ ∑ hu , s ,n , m ( t ) δ ( τ −τ n , i ) +∑ hu , s ,n ( t ) δ ( τ −τ n )
huNLOS NLOS NLOS

n=1 i =1 m n=3
(2)

N
H u , s ( f , t )=∑ hu , s (τ , t)e− j 2 π τ f n

n=1 (3)

To calculate the spatial-temporal correlation function of the 2 × 2 antenna system,

the time average operator is employed, as shown in:

ρ( Δd s , Δd u , Δt , Δf =0)=⟨ H u , s ( f ,t ) × H ¿u , s ( f , t+ ∆ t ) ⟩ =¿
1 1 2 2

√
N M
Pn
∑ ∑ ¿¿
M m=1 (4)
n =1

The cross-spatial correlation function of the channel at the transmitter, with Δ𝑡 =

Δ𝑓 = 0 and Δ𝑑𝑢 = 0, can be expressed as:

√
T
N j 2 π (^r tx , n ,m ×∆ dtx , u)
Pn M
ρ ( ∆ d s )=∑ ∑e λ0
M m=1 (5)
n=1

7
 When Δ𝑡 = Δ𝑓 = 0 and Δ𝑑𝑠 = 0, the cross spatial correlation function at the
receiver is as follows : Similarity

√
T
N j 2 π ( r^rx , n ,m ×∆ drx , u)
Pn M
ρ ( ∆ d u )=∑
M∑
λ0
e (6)
n=1 m=1

2.2. The correlation properties in BS side:

Figure 2. The correlation properties in BS side [2]

Based on Figure 2, the measured distance of the antenna components on the BS side is
shown. The UMi and RMa systems appear to be relatively stable, with the first amplitude
of UMi measuring at 17.8, followed by 17.7 and 17.5. The RMa's amplitudes are 14, 13.9
and 13.8, respectively. However, the Indoor scenario displays a downward trend, with its
maximum value being close to UMi's, but subsequent values decreasing to 16, 13.9, 10
and ultimately about 7, as illustrated in the graph. The first minimum correlation values
are about 0.0051 in Indoor, 0.007 in UMi and 0.011 in RMa. The second and third
minimum correlation values are respectively 0.016 and 0.027 in Indoor, 0.022 and 0.036
in UMi, and 0.032 in RMa. Based on the data presented in Figure 2, it can be inferred that
UMi exhibits the best signal quality due to its consistently high amplitude. And we can
conclude that increasing the antenna element’s spacing in BS ( Δ d s) will reduce the
amplitude of the correlation properties.

8
 2.3. The correlation properties in MS side:

Figure 3. The correlation properties in MS side [3]

Based on Figure 3, the highest amplitude is observed in UMi, with a value over 18.
The next highest amplitude is seen in Indoor with a value of approximately 17, while the
lowest is observed in RMa at nearly 14. All scenarios experience a reduction in
amplitude, but the RMa scenario shows the least change, with amplitudes of 13.6 and 10.
The Indoor scenario has amplitudes of 13.5 and 7, while the UMi scenario has values of
11 and 6.5. The minimum correlation value for RMa falls within the range of 0.005 to
0.01, with a value of 0.098. For Indoor and UMi scenarios, the minimum correlation
values are approximately 0.006 and 0.011, respectively. Therefore, we can conclude that
the RMa is the most stable, next is the Indoor and the last is UMi. Similarly to the BS
side, if we increase the antenna element’s spacing in MS ( Δ d u), it will reduce the
amplitude of the correlation properties.

3. CONCLUSION
From figures 1 and 2, it can be inferred that the MS is relatively more stable
compared to the BS, because the distance between the first maximum and minimum
values are relatively close to each other. Additionally, any alterations made on the
transmitter side will have a greater impact on the receiver side in a BS system. In

9
contrast, the MS is designed to be adaptable to mobile conditions, which enhances its
flexibility.

4. REFERENCES
[1] ETSI TR 138 901 V14.0.0 (2017-05), ”5G -Study on channel model for
frequencies from 0.5 to 100 GHz” (3GPP TR 38.901 version 14.0.0 Release 14)
[2] Thu Nga Nguyen, Van Duc Nguyen, “A performance comparison of
theSCM and the Onering channel modeling method for MIMO-OFDMAsystems”, in
Wireless Communications MobileComputing, Volume 16,Issue 17, 2016.
[3] Samimi, M. K., Sun, S., Rappaport, T. S, “MIMO channel modelingand
capacity analysis for 5G millimeter-wave wireless systems”, 10thEuropean Conference
on Antennas and Propagation (EuCAP) 2016,doi:10.1109/eucap.2016.7481507.
[4] Yawei Yu, Peter J. Smith, Pawel A. Dmochowski, Jianhua Zhangand
Mansoor Shafi, “3D vs. 2D Channel Models:Spatial Correlationand Channel Capacity
Comparison and Analysis”, IEEE ICC WirelessCommunications Symposium, 2017.

10