MIMO Antenna System

1
 Evolution of Wireless Networks at a Glance

2.5G/3G/3.5G 4G
2G
1G 5G
Hierarchal Cell Smart Antennas
Digital Modulation
Structure, MIMO Mm-wave Technology
Analog Convolution Coding
Turbo-coding Adaptive System Massive MIMO
Power Control
OFDM Modulation
AMPS, TACS PDC, GSM EDGE, UMTS/HSPA
GPRS LTE

AMPS: Advanced Mobile Phone System,  LTE standard targets higher peak data rates,
TACS: Total Access Communication Systems, higher spectral efficiency, lower latency, flexible
PDC: Personal Digital Cellular (Japan), channel bandwidths, and system cost compared
GSM: Global System for Mobile to its predecessor UMTS/HSPA
Communications,
GPRS: General Packet Radio Service,  LTE also referred to as MAGIC: Mobile
EDGE: Enhanced Data GSM Environment, multimedia, Anywhere anytime, with Global
UMTS: Universal Mobile Telecommunication mobility support, Integrated wireless solution,
System, and Customized personal service.
HSPA: High Speed Packet Access,
LTE: Long Term Evolution

6G, 7G and Beyond in Future……

Reference: L. Song and J. Shen, “Evolved Cellular Network Planning and Optimization for UMTS and LTE,” CRC Press, 2011. 2
 1G:- 1’st Generation
 1G refers to 1st generation of mobile telecommunication
 It is developed in 1980s and completed in early 1990s.
 It provides a speed up to 2.4kbps.
 It is based on analog system.
 It allows user to make call in one country.
 It has low capacity, unreliable handoff, poor voice links, and no
security at all since voice calls were played back in radio towers,
making these calls susceptible to unwanted eavesdropping by third
parties.

3
 2G :- 2’nd Generation

 2G refers to 2nd generation of mobile telecommunication.

 It was developed in late 1980s and completed in late 1990s.
 It is based on digital system.
 It provides a speed of up to 64 kbps.
 It provides services like voice and SMS with more clarity.
 Major prominent technologies were GSM and CDMA.

4
 3G :- Third Generation
 NTT DoCoMo launched the first commercial 3G network on 1 October
2001, using the WCDMA technology
 Bandwidth of 3G network is 128 Kbps for mobile stations, and 2 Mbps
for fixed applications.
 The current trend in mobile systems is to support the high bit rate data
services at the downlink via High Speed Downlink Packet Access
(HSDPA).

5
 4G: 4th Generation
 It was developed in the year 2010.
 It is faster and more reliable.
 It provides speed up to 100 Mbps.
 It provides high performance like uploading and downloading speed.
 It provides easy roaming as compared to 3G.
 Use of a higher Layer Protocol (IP) as transport medium affords
intelligence at every stage within the network relative to a service

6
 5G: 5th Generation
 It is the next major phase of mobile telecommunication & wireless
system.
 It is 10 times more faster than 4G.
 It has a expected speed of 1Gbps.
 Lower cost than the previous version.
 Commercially launched in 2019.

7
 Some Future Predictions on Wireless Traffic
● Monthly global mobile data traffic will surpass 77 exabytes by 2022.

1 Exabyte = 1000 Petabyte = 106 Terabyte = 250 million DVDs

● The number of mobile-connected devices exceeded the world’s population in 2014.

CAGR: Compound Annual Growth Rate

https://disruptive.asia/5g-mobile-data-growth-2020-cisco-vni/ 8
 Characteristics of Different Cellular Systems
Parameters 2.5G 3G 3.5G 4G
EDGE CDMA2000 UMTS EV-DO HSDPA EV-DV LTE

Channel 0.2 1.25 5 1.25 5,10 1.25, 3.75 5,10,15,20

Bandwidth (MHz)
Duplexing FDD FDD FDD FDD FDD FDD FDD/TDD
Multiplexing TDMA TDMA WCDMA TD-CDMA WCDMA TD- OFDM/SCF
CDMA DMA
Modulation GMSK/ GMSK/ QPSK QPSK/8PSK/ QPSK/ QPSK/8PS QPSK/
8PSK 8PSK 16QAM 16QAM K/16QAM 16QAM/
64QAM
Coding C CTC CTC CTC CTC CTC CTC
Maximum Data 0.04 0.05 0.14 1.8 2 1 50
Rate in Uplink
(Mbps)
Maximum Data 0.18 0.38 0.38 3.1 7.2 3-5 100
Rate in Downlink
(Mbps)
GMSK: Gaussian minimum shift keying, QPSK: Quadrature phase shift keying, QAM: Quadrature amplitude modulation,
TD-CDMA: Time division-synchronous CDMA, OFDMA: Orthogonal frequency division multiple access,
SC-FDMA: Single carrier frequency division multiple access, CTC: Convolutional / Turbo coding

L. Song and J. Shen, “Evolved Cellular Network Planning and Optimization for UMTS and LTE,” CRC Press, 2011. 9
 MIMO: Major Enabling Technology 4G/5G Networks
• Use of Multiple-input-multiple-output (MIMO) systems
• Use of Adaptive Beam Forming
Various Enabling • Use of Advanced Modulation and Coding schemes (AMC)
Technologies • Use Multiple Access Techniques
• Use of smaller cells (pico/femto)
• Use MM-waves

 Multiple antennas are used at both the transmitter and receiver to increase wireless channel
capacity.
 Different signals are transmitted out of each antenna simultaneously in the same bandwidth
and then separated at the receiver.
 It can support multiple independent channels in the same bandwidth, if the multipath
environment is rich enough. 10
Reference: L. Song and J. Shen, “Evolved Cellular Network Planning and Optimization for UMTS and LTE,” CRC Press, 2011.
 Shannon–Hartley Theorem
The Shannon–Hartley theorem states the channel capacity C, meaning the theoretical
tightest upper bound on the information rate of data that can be communicated at an
arbitrarily low error rate using an average received signal power S through an analog
communication channel subject to additive white Gaussian noise of power N

 S
C '  B log 2 1  
 N
where
C’ is the channel capacity in bits per second, a theoretical upper bound on the net bit rate
(information rate) ;

B is the bandwidth of the channel in hertz (passband bandwidth in case of a bandpass signal);

S is the average received signal power over the bandwidth, measured in watts

N is the average power of the noise and interference over the bandwidth, measured in watts

S/N is the signal-to-noise ratio (SNR) or the carrier-to-noise ratio (CNR) of the communication
signal to the noise and interference at the receiver (expressed as a linear power ratio, not as
logarithmic decibels).
11
If the SNR is 20 dB, and the bandwidth available is 4 kHz, which is appropriate for
telephonic communications,

S/N = 100 = 20 dB.

 S
C '  4 log2 1  100  C '  26.63 kb/s
C '  B log 2 1  
 N

If the requirement is to transmit at 50 kbit/s, and a bandwidth of 10 kHz is used,

then the minimum S/N required is given by 50000 = 10000 log2(1+S/N)

SNR =14.91 dB.

12
 Shannon’s Channel Capacity Theorem: SISO
SISO  Single Input Single Output
Here t = Time instant, τ = Delay

If the channel can be simplified as being time and frequency invariant, the dependence of t
and τ is dropped such that the channel is simply denoted by a scalar quantity
ht ,   h y  hx  n
y  Received Signal
x  Transmitted Signal
n  Additive White Gaussian Noise (AWGN) with zero mean and variance  n2

In a noise-limited scenario, the spectrum  PT 2 

efficiency of a channel is upper bounded by C SISO 
 log 2 1  2 h 
the Shannon capacity C ( = CSISO here)  n 
h  Power gain of the channel
2

Unit of C is bits/sec/Hz
PT  Transmit Power

[R1] R. Tian, “Design and Evaluation of Compact Multi-antennas for Efficient MIMO Communications,” Doctoral Dissertation,
2011, Lund University. Link: http://lup.lub.lu.se/record/2199615. 13
 Shannon’s Channel Capacity Theorem in SISO System

y  hx  n
 P 2
C  log 2 1  T2 h 
 n 
The expression for Channel capacity indicates that channel capacity only increases logarithmically with
an increase in transmit power

PT High signal-to-noise ratios (SNRs)  PT   n2

SNR 
 n2
 P 2
C  log 2  T2 h   2 log 2 h  log 2 SNR 
n 
In high SNR case, a 3 dB increase (or doubling) of the power would only yield 1 extra
bits/s/Hz.
Thus the performance of a wireless communication system is constrained by either power-limited
operation or bandwidth-limited operation. Given a frequency bandwidth, the use of multiple
antennas can improve the spectrum efficiency of the system.
[R1] R. Tian, “Design and Evaluation of Compact Multi-antennas for Efficient MIMO Communications,” Doctoral Dissertation, 2011,
14
Lund University. Link: http://lup.lub.lu.se/record/2199615.
 MIMO System: Sampled Vector Signal Model
M T  Number of Transmitters
M R  Number of Receivers
y  Hx  n
x  Transmitted signal vector
y  Received signal vector
n  AWGN vector at the receive antennas
H  Channel matrix

 h1,1 t ,  h1, 2 t ,  ... h1, M T t ,  

Matrix Dimensions  h t ,  h t ,  ... h t ,  
x M T 1  2,1 2, 2 2, M T 
 . . . . 
y M R 1 H 
. . . .
n M R 1  
 . . . . 
H M R  MT hM ,1 t ,  hM , 2 t ,  ... hM , M t , 
 R R R T 
[R1] R. Tian, “Design and Evaluation of Compact Multi-antennas for Efficient MIMO Communications,” Doctoral Dissertation, 2011,15
Lund University. Link: http://lup.lub.lu.se/record/2199615.
 MIMO Channel Capacity
y  Hx  n  PT 2 
C SISO 
 log 2 1  2 h 
H  Channel matrix  n 
Elements
Scalar SISO channel between:
hi , j t ,  i th Receive Antenna
j th Transmit Antenna

The capacity of an instantaneous MIMO I M R  M R  M R  Identity Matrix

channel:
detA  Determinant of A  A
 PT H 
C MIMO  log 2 det  I M R  HH   AH  Hermitian / Conjugate Transposeof A
 M T n
2

PT  Transmit Power  A H
 A  T *

• In this expression for CMIMO, the transmit power PT is assumed to be equally

allocated over the MT transmit antennas, corresponding to the case of no channel state
information (CSI) at the transmitter.

[R1] R. Tian, “Design and Evaluation of Compact Multi-antennas for Efficient MIMO Communications,” Doctoral Dissertation, 2011,16
Lund University. Link: http://lup.lub.lu.se/record/2199615.
 MIMO Capacity versus SISO Capacity
 h1,1 t ,  h1, 2 t ,  ... h1, M T t ,  
 h t ,  h t ,  ... h t ,  
 2,1 2, 2 2, M T 
 . . . . 
H 
. . . .
 
 . . . . 
hM ,1 t ,  hM , 2 t ,  ... hM , M t , 
 R R R T 

 PT H 
C MIMO  log 2 det  I M R  HH 
 M T n
2

• Assumption: channels for all pairs of RX and TX antennas are mutually uncorrelated
independent and identically distributed (IID) complex Gaussian random variables

CMIMO  MCSISO

The full potential of MIMO systems relies on the properties of the overall propagation
channel H, which includes the impacts of the TX and RX multi-antenna systems. 17
 SISO  MIMO: Increase in Spectral Efficiency
 PT 2   PT H 
C SISO  log 2 1  2 h  C MIMO  log 2 det  I M R  HH 
 n   M T n
2


Figure : Tradeoff in required bandwidth BW

and number of antennas (M) in achieving
1Gbps for a M × M MIMO channel with 20
dB SNR.

Trade-off between the required bandwidth (BW) and the number of antennas in achieving a
target date rate of 1 Gbps
Let BW for SISO case be 150 MHz for SNR of 20 dB

C  150 106 log2 1  100   1 Gbits/sec 18

 SISO  MIMO: Increase in Spectral Efficiency
Let BW for SISO case be 150 MHz for SNR of
20 dB

On the other hand, for the same SNR, assuming a MIMO system of MR = MT = 6, the
occupied bandwidth can be approximately reduced to 30 MHz, if the aforementioned
IID channel is assumed;

However, practical wireless channels may not be IID  slight increase in occupied
bandwidth in MIMO case; but still the required bandwidth will be much smaller than SISO
case

This simple example demonstrates the effectiveness of MIMO technology in drastically

improving spectrum efficiency, which is especially beneficial due to the frequency spectrum
being a scarce and expensive resource today. 19
SISO  MIMO: Increase in Spectral Efficiency

20
 
CSISO  log 2 1  SNR h
2
 CMIMO

 log 2 det  I M R 

SNR
MT
H 
HH 


 2
 SNR M T 2 

MR
CSIMO  log 2 1  SNR  hi  CMISO  log 2 1  hi 
 i 1   MT 1 
hi is gain of ith receiver antenna hi is gain of receiver antenna with ith 21
transmitter antenna
 Multipath Propagation
• In free space, electromagnetic waves that are launched from the TX antenna reach the RX
antenna along the line-of-sight (LOS) propagation path.

• However, in mobile communications, the key propagation mechanism is non-LOS

(NLOS).

• non-LOS (NLOS) multi-path propagation  mainly caused by the interaction of radio

waves with different types of scattering objects in the radio channel;

22
 Multipath Propagation
• In a SISO system, one needs to design dedicated transceiver algorithms in order to combat
performance degradation due to multi-path fading

• However, multi-path fading turns out to be the key for utilizing multi-antenna
techniques;

•For instance, there is a significantly smaller probability that all the signals across spatially
separated antennas experience deep fade simultaneously, as compared to it occurring in the
signal from one antenna  Space-diversity techniques

23
 NLOS / LOS Scenario: SISO Case
• In a NLOS scenario, there are a large
number of scatterers that cause multipath
propagation, but there is no dominant path.

• In a SISO channel, the complex-valued baseband representation of the channel response h

can be modelled as a zero mean complex Gaussian variable as follows:


h ~ CN 0,  02 
Mean  0, Variance   02

  02  Real-valued Gaussian (or

Re h  and Im h  ~ N  0,  normal) distribution
 2 

In this case, the amplitude |h| is Rayleigh distributed and the phase φ is uniformly
distributed between 0 and 2π
24
 NLOS / LOS Scenario: SISO Case

• On the other hand, when a dominant path of amplitude A0 exists among the Multi-path
components (MPCs), the fading channel is best modelled using a Rician distribution.

• The scenario is commonly referred to as LOS, although the dominant component does
not necessarily propagate along the line-of-sight path.

• Rician K-factor is defined as the ratio of the power in the dominant component to the
power in the scattered components:

A02
K K  0 : Rician  Rayleigh
2 02

25
 NLOS / LOS Scenario: MIMO Case
 h1,1 t ,  h1, 2 t ,  ... h1, M T t ,  
 h t ,  h t ,  ... h t ,  
 2,1 2, 2 2, M T 
 . . . . 
H 
. . . .
 
 . . . . 
hM ,1 t ,  hM , 2 t ,  ... hM , M t , 
 R R R T 

• The independent and identically distributed (IID) Gaussian model used in the
mathematical formulation assumes that each element of the channel matrix H is an IID
variable of 
CN 0,  02
Therefore, such a H-matrix is referred to as the IID Rayleigh MIMO channel.

• The rows or columns of such a channel matrix are linearly independent, which ensures that
the channel matrix is full rank.

• The full rank condition of the channel in NLOS scenario is favourable for MIMO systems.

• On the other hand, the dominant component in the LOS scenario can impair this condition
by causing the condition number of H to increase, due to large amount of power being
available to only one sub-channel (i.e., the dominant path) 26
 Analytical Channel Models
In order to include spatial correlation properties into the channel model, the the correlation
at the TX and RX needs to be added separately.

The composed channel matrix  H K  R H IID R

1/ 2
R  1/ 2 T
T

TX and RX sides are flipped to make the system look similar to the channel matrix
27
 Performance Metrics of MIMO Antenna Systems

Standard Antenna Parameters:

• Reflection Coefficient or S11: Operating Frequency, Impedance Bandwidth
• Input Impedance Profile  Equivalent Circuit Modeling of Antennas
• Radiation Patterns (3D, 2D)
• Directivity/Gain
• Radiation Efficiency
• Axial Ratio (Polarization)

Some Additional Parameters for MIMO Antenna:

• Envelope Correlation Coefficient (ECC or ρe)
• Channel Capacity Loss (CCL)
• Total Active Reflection Coefficient (TARC)
• Mean Effective Gain (MEG)

• ECC and CCL can be evaluated from both S-parameters and 3D radiation patterns
• TARC is evaluated from S-parameters only
• MEG is evaluated from 3D radiation patterns only
28
 Envelope Correlation Coefficient (ECC or ρe)
 ECC or ρe describes how much the communication channels are isolated or correlated
with each other.
 This metric considers the radiation pattern of the antenna system, and how much the
patterns affect one another when operated simultaneously
 Let us focus on Uniform Propagation environment only with balanced polarization

Complex Correlation  ij   Ei  E*    E  E*  d

j i j

Coefficient  E   2  E   2 d  E   2  E   2 d

  i i
    j j 


Envelope Correlation 2  E E   E E d

* *
2

 e,ij   ij 
i j i j

 d
  E   E  d  E   E 
Coefficient 2 2 2 2

i i j j


e  jkr E   E d

2
Ei r ,    Ei r , ,   
r
Ei  
 e,ij 
 i j

 E  d E  d

2 2

Ei    E i  ,  ˆ  Ei  ,  ˆ

i j

 Hermitian Product Operator

29
 Orthogonally placed Electric Dipole

y
x
xy plane yz plane

y
x
xy plane yz plane
30
 Orthogonally placed Electric Dipole

z z z

y y y
x x x

31
 Correlation in MIMO Systems
 h1,1 t ,  h1, 2 t ,  ... h1, M T t ,  
 h t ,  h t ,  ... h t ,   • Although IID Rayleigh Channels are more
 2,1 2, 2 2, M T  desirable, correlations do often exist among the
 . . . . 
H  elements of the channel matrix in practical cases
. . . .
 
 . . . . 
hM ,1 t ,  hM , 2 t ,  ... hM , M t , 
 R R R T 

• An antenna is characterized by its far-field radiation / gain pattern denoted by:

E , ()  E  ˆ  E  ˆ

2
G , ()  E , ()   Solid Angle
2 

  G ( ,  )  G ( ,  )sin dd  4

0 0

Here E  and E  are normalized value of electric field distribution in space
normalization has been done with respect to isotropic antenna
E    - polarized componentof incident field Pavg,  average incident power of  - polarized field
E    - polarized componentof incident field Pavg ,  average incident power of  - polarized field
• The power ratio between the two polarizations  Cross-polarization discrimination (XPD)

Pavg ,
XPD or  
Pavg ,

• The propagation channel is characterized by the incident field on the RX antenna system.
This can be described by a distribution of the angular power spectrum (APS):

 , ()    ˆ    ˆ

 ,   probability density function for incident electromagnetic fields

2  2 

   ( ,  ) sin dd     ( ,  ) sin dd 1

0 0 0 0

33
 Correlation in MIMO Systems
As reference scenario, generally a uniform 3D distribution is used to model a completely
random environment for angular power spectrum (APS):

Uniform
, ()  1

For uniform 3D environment with balanced polarization i.e. χ = 1, the complex

correlation coefficient can be defined as below.

The complex correlation coefficient between the i-th and j-th antenna elements can
be calculated with the following expression using their radiation patterns as:

 E          d
* *
 E E E
 ij  i j i j

 E   2  E   2 d  E   2  E   2 d
  i i
    j j 

More discussion
on this later
Ei , Ei    and  polarized far - field componentsof the i - th antenna

AB *
ij 
AB 34
 Correlation in MIMO Systems
Pavg ,
 , ()    ˆ    ˆ  Given Propagatio n Scenario (say) 
Pavg ,
The complex correlation coefficient

  
 E  
  1   i  j 
 E *
    
1
Ei  E *
j     d
1  
 ij 
Ge ,i Ge , j
Ei , Ei    and  polarized far - field componentsof the i - th antenna

The normalization factor Ge denotes the mean effective gain (MEG) of the antenna:

  
Gi      Gi   d
1
Ge,i   
1  1  
Gi    Ei  
2
Gi    Ei  
2

35
 APS for different scenario and Correlation
As reference scenario, generally a uniform 3D distribution is used to model a completely
random environment:
PUniform
, ()  1
For uniform 3D environment with balanced polarization i.e. χ = 1, the complex
correlation coefficient is given by:

 ij   Ei  E*    E  E*  d

j i j

 E   2  E   2 d  E   2  E   2 d
  i i
    j j 

However for many practical situations, a Gaussian distribution (truncated), is a statistically
appealing form for angular power spectrum (APS):

 ,
Gaussian

( ,  )  exp 

    0 2

   0 2 


 2  
  2 
2 2

Mean angle-of-arrival (AOA)   0 , 0 

The standard deviation of the angular spread    ,  
Generally the performance converges to that of the uniform 3D distribution when the
36
angular spread is large enough.
 Performance Metrics of MIMO Antenna Systems

Standard Antenna Parameters:

• Reflection Coefficient or S11: Operating Frequency, Impedance Bandwidth
• Input Impedance Profile  Equivalent Circuit Modeling of Antennas
• Radiation Patterns (3D, 2D)
• Directivity/Gain
• Radiation Efficiency
• Axial Ratio (Polarization)

Some Additional Parameters for MIMO Antenna:

• Envelope Correlation Coefficient (ECC or ρe)
• Channel Capacity Loss (CCL)
• Total Active Reflection Coefficient (TARC)
• Mean Effective Gain (MEG)

• ECC and CCL can be evaluated from both S-parameters and 3D radiation patterns
• TARC is evaluated from S-parameters only
• MEG is evaluated from 3D radiation patterns only
37
 Envelope Correlation Coefficient (ECC or ρe)
 ECC or ρe describes how much the communication channels are isolated or correlated
with each other.
 This metric considers the radiation pattern of the antenna system, and how much the
patterns affect one another when operated simultaneously
 Let us focus on Uniform Propagation environment only with balanced polarization

Complex Correlation  ij   Ei  E*    E  E*  d

j i j

Coefficient  E   2  E   2 d  E   2  E   2 d

  i i
    j j 


Envelope Correlation 2  E E   E E d

* *
2

 e,ij   ij 
i j i j

 d
  E   E  d  E   E 
Coefficient 2 2 2 2

i i j j


e  jkr E   E d

2
Ei r ,    Ei r , ,   
r
Ei  
 e,ij 
 i j

 E  d E  d

2 2

Ei    E i  ,  ˆ  Ei  ,  ˆ

i j

 Hermitian Product Operator

38
 Envelope Correlation Coefficient (ECC or ρe)
e  jkr
Ei r ,    Ei r , ,    Ei   E   E d
2

r  e,ij 
 i j

 E  d E  d

2
E    E  ,  ˆ  E  ,  ˆ
2
i i i i j

r2
U i  ,    Ei r , ,    Ei  ,    Average radiation intensity
2 1 2

2 2
4U i  ,   4 1 2 4 1 2
Di  ,      2 Ei    e P  2 Ei  
Prad ,i Prad ,i   i in  
Prad ,i  ei Pin ; ei  rad ,i  Radiation efficiency; Pin  Input Power

 2 Di  ,    ei Pin Di  ,   Maximum value of this can

Ei   
2
 ei Pin  
 4  2 used for normalization.


Dmax, i  Fi ( ,  )  Normalized 3D radiation pattern
Ei    ai Fi  ,   of the antenna when ith port is excited
2
a i  Excitation amplitude of ith port when it is
excited 39
 Derivation of ECC formula from S-parameters
Let two antennas (i = 1 and j = 2) be driven by two
Antenna-1 generators at the same frequency.
1. Antennas not necessarily well-matched
2. Antennas have nonzero mutual coupling
Thus S11, S12, S21 and S22 all are considered non-zero

E    E d
2

 e,12 
 1 2

 E  d E  d

2 2
Antenna-2
1 2

D1  D2 
E1    a1 F1  ,   E 2    a2 F2  ,  
2 2
Dmax, 1  D1 Dmax, 2  D2

Total radiated electric field due to the two antennas: Etot   E1   E2 

Etot   d  E1    E 2   d
1 2 1 2
 
The total power
Prad ,tot 
radiated by both 2 2

 
antennas in whole
space: 
1
 1
E    2
 E   2
 E1    E 2    E 2    E1   d
2
2
40
 Derivation of ECC formula from S-parameters
Prad ,tot 
1
2  
E1   2
 E 2   2
 E1    E 2    E 2    E1   d 
D1  D2 
E1    a1 F1  ,   E 2    a2 F2  ,  
2 2
1  D1 2   
F1  ,   d
D
1
   
2

2

2 
   1  1     R11  1
2 2
E d  a F , d  R a
2  2 4
1 11 1

1  D2   
F2  ,   d
D
1
   
2

2

2 
   2  2     R22  2
2 2 2
E d  a F , d  R a
2  2 4
2 22 2


1  D1 D2 *   
1
2 
E1    E 2  d  
2  2 2
a1 2
a F1  ,   F2  ,   d 
  R a a
21 1 2
*


1  D2 D1   
1
2 
E 2    E1  d  
2  2 2
a 2 1
a *
F2  ,   
 F1  ,   d 
  R a 
a
12 2 1
*

   
   
D1 D2 D2 D1
R21  F1  ,    F2  ,   d R12  F2  ,    F1  ,   d
4 4
41
 Derivation of ECC formula from S-parameters
Prad ,tot 
1
2  
E1   2
 E 2   2

 E1    E 2    E 2    E1   d

Prad ,tot  R11 a1  R a a  R a a  R22 a2

2 * * 2
12 2 1 21 1 2

 a1 
a  R
 R11 R12  In matrix notation: Prad ,tot  a H Ra
 a2   R21 R22 
Here H in the superscript denotes Hermitian Operator

Prad ,tot  a Ra  a
H
 1
*
a2
*
  R11
R
R12   a1 
R22  a2 
 21

Prad ,tot  a Ra  a
H
 1
*
a2
*
  R11 a1  R12 a2 
R a  R a 
 21 1 22 2 
 Derivation of ECC formula from S-parameters
Prad ,tot 
1
2 

E1   2
 E 2   2
 E1    E 2    E 2    E1   d 
E1    a1
D1 
F1  ,  
2

Prad ,tot  R11 a1  R12 a1a2*  R21 a2 a1*  R22 a2
2 2
F1  ,   d
D 2
R11  1
4 
 a1 
a  R
 R11 R12  In matrix notation: Prad ,tot  a H Ra
 a2   R21 R22 
Here H in the superscript denotes Hermitian Operator

 E1  d  2R11 a1  E2  d  2R22 a2

2 2 2 2

 E1   E2 d   E2   E1 d

*
Now, using the property of Hermitian Product:

 E   E d 
  E1   E2 d E2   E1 d  2R21a1a2* 2R12 a2 a1* 
2
1 2

 E   E d
2
4 2 R12 R21 a1 a2
2 2
i j R12 R21
 e,ij   
 E  d  E j   d 4 2 R11 R22 a1 a2
2 2 2 2
R11 R22
i
43
 Derivation of ECC formula from S-parameters
 a1   b1   R11 R12 
a  b  R Prad ,tot  a H Ra
R22 
Antenna-1
 a2  b2   R21
The total radiated power for this two-antenna system can
be expressed in terms of S-parameters as follows:
2 2
Prad ,tot   a   bi2  a H a  b H b
2
 S11 S12 
i
S
S 22 
Antenna-2 i 1 i 1
 S 21

 a H I  S H S a  a H Ra 
For two-port case

1 0  S11*
R  I S S    *
*
S 21   S11 S12   1  S11  S 21

2 2

 S11* S12  S 21
*
S 22  
2 
H
* 

0 1  S12 S 22   S 21


S 22   S12* S11  S 22
*
S 21  1  S12  S 22 
2


R21   S S11  S S 21   S S  S S 22
*
12
*
22   *
11 12
*
21  *
 R12*
2
R12 R21 R12
2
S S12  S S 22 * *

e   
  
11 21

1  S11  S 21 1  S12  S 22
2 2 2 2
12
R11 R22 R11 R22 44
 Summary of ECC Formula for two antenna system

E    E d
2 2

 S S12  S S 22
* *

e 
1  S 1  S 
1 2 11 21
 e,12 
 S 21  S 22
2 2 2 2

 E  d E  d

2 12
2
11 12
1 2

• Uniform multipath environment of • Requires knowledge of S-parameters only

balanced polarization is considered
• Can be evaluated for wide frequency bands
• Requires 3D radiation-pattern
measurements and numerical • Not accurate as a number of approximations
integration are made  antenna radiation efficiency and
internal loss are neglected
• Solved for only one frequency:
Effective for narrowband formulation

ECC < 0.5 has been set as an acceptable value for 4G wireless systems. However generally
designers look for ECC < 0.1

45
Estimating Channel Capacity from S-parameters
Correlation matrix R can be
expressed in terms of S-matrix as
(assuming lossless scenario):
R  I  SH S
Thus for TX side: R T  I  STH ST

RX side: R R  I  S H
R SR

MIMO Channel capacity:


C MIMO  log 2 I M R  HH H
MT

PT
  SNR ‘||’  Used to
 2
n denote determinant
We take Kronecker channel model as:
H  R H IID R
1/ 2
R  T 
1/ 2 H

Let: M R  M T  M Also let us denote: H IID  HW 46

 Estimating Channel Capacity from S-parameters
H  R HW R
1/ 2
R  T 
1/ 2 H

PT
 2
 SNR
n


CMIMO  log 2 I M  HH H
M
For high SNR conditions

CMIMO  log 2 HH H
M R  MT  M M

C MIMO  log 2

M
1/ 2
R HW R
R  T 
1/ 2 H
R 1/ 2
R 
HW R T 
1/ 2 H
H


 log 2
M

R1R/ 2 HW R1T/ 2 
H
R1T/ 2 HWH R1R/ 2  H


 log 2
M
1/ 2
R HW R T H R
R
H
W  1/ 2 H
R 

 log 2 HW HWH  log 2 R T R R
M 47
  CMIMO  CMIMO,ideal  CLoss ,T  CLoss , R
CMIMO  log 2 HW HWH  log 2 R T R R
M

CMIMO ,ideal  log 2 HW HWH
M
Thus we can define a parameter called Channel Capacity
Loss (CCL or CLoss) having units bits/s/Hz as:


C Loss ,T   log 2 RT   log 2 I  STH ST  The negative sign is taken in the same
convention as that of Return loss and S11


C Loss , R   log 2 RR   log 2 I  S HR S R 
For the 2-element antenna system the matrix RR can be evaluated as:
 1  S
2
 S
2
S11* S12  S 21
*
S 22 
RR  I  SR SR   *
H 11 21
2
 S12 S11  S 22 S 21 1  S12  S 22 
* 2

After computing the determinant of the matrix RR , CCL can be estimated

48
 Channel Capacity Loss for M × M MIMO System

CMIMO  CMIMO,ideal  CLoss ,T  CLoss , R


CMIMO ,ideal  log 2 HW HWH
M


C Loss ,T   log 2 RT   log 2 I  STH ST 
M R  MT  M C Loss , R   log 2 RR   log 2 I  S H
R SR 
• Thus due to the correlation in both the TX and RX ends, there is a degradation in the
MIMO channel capacity that would have been obtained in ideal scenario

• For Antenna design, we generally design either for TX end or RX end  CCL can be
quickly characterized from either S-parameters

• More accurate CCL can be obtained by evaluating the complex correlation coefficient
matrix from radiation patterns

CCL< 0.5/bits/sec/Hz has been set as an acceptable value for 4G wireless systems 49
 Total Active Reflection Coefficient (TARC)
The TARC is defined as the ratio of the square root of the total reflected power divided by
the square root of the total incident power in a multi-port antenna system.

TARC for an M-port antenna system: M For evaluating TARC

 bi
2
generally incident signals of
Here ai and bi are the incident and unit amplitude and random
i 1
reflected signals at ith antenna port at  phase are considered:
M

 ai
2
Clearly: 0   1t
a
ai  1i
i 1

Let us consider a two-port  a1   b1   S11 S12 

a  b  S b  Sa
antenna system: a  2 b2   S 21 S 22 

b1  b2 S11 a1  S12 a2  S 21 a1  S 22 a2
2 2 2 2

at  
a1  a2 a1  a2
2 2 2 2

Let us consider one incident signal as reference and the other a1  10, a2  1
having a random phase θ, that has to be swept between 0 to π 50
 Total Active Reflection Coefficient (TARC)

j 2 j 2
S11a1  S12 a2  S 21a1  S 22 a2
2 2
S11  S12 e  S 21  S 22 e
 
t
a 
a1  a2
2 2
2

a1  10, a2  1

• By varying θ, designers investigate the effect of the random phase variation between the
two ports on the resonance behaviour and effective bandwidth of the TX / RX MIMO
antennas

• θ is generally taken as i.i.d. Gaussian random variable because MIMO channels are
assumed as Gaussian and multipath spread in the propagation channel.

• TARC for MIMO antennas having M > 2 can also be evaluated in the same manner, only
there are multiple random phase variables θi

• It is often advised to take multiple random variations of θ and compute TARC for all the
cases, and then compute the average TARC which gives a better measure
51
 Mean Effective Gain (MEG)
  
G      G    d
1
Ge   
1  1  
Pavg ,
Cross Polarization Discrimination  
Pavg ,

Statistical distribution P    P  ,  ,
Antenna Gain Components G    G  ,  ,
P   P  ,  
of incoming waves in
along θ and φ directions G    G  ,   the environment
2 

The conditions   G ( ,  )  G ( ,  )sin dd  4

0 0
which must be
satisfied: 2  2 

  P ( ,  ) sin dd    P ( ,  ) sin dd 1

0 0 0 0

This numerical method allows us to get MEG using the simulated/measured gain patterns
in an ideal environment (i.e., the simulation tool or an anechoic chamber), and a model
of the environment suitable for the application for which the antenna is being designed.
General requirement for 4G LTE MIMO systems: |MEGi | / |MEGj|  1 52
53
 References
[R1] Riuyuan Tian, “Design and Evaluation of Compact Multi-antennas for Efficient MIMO
Communications,” Doctoral Dissertation, 2011, Lund University.
[R2] J. Winters, “On the capacity of radio communication systems with diversity in a Rayleigh fading
environment,” IEEE Journal on Selected Areas in Communications, vol. 5, pp. 871 – 878, June 1987.
[R3] G. J. Foschini and M. J. Gans, “On limits of wireless communications in a fading environment when
using multiple antennas,” Wireless Personal Communication, vol. 6, pp. 311 – 335, 1998.
[R4] I. E. Telatar, “Capacity of multi-antenna Gaussian channels,” European Transactions on
Telecommunications, vol. 10, pp. 585 – 595, 1999.
[R5] D. S. Shiu, J. Foschini, J. Gans, and J.M Kahn. 2000. “Fading correlation and its effect on the
capacity of multi-element antenna system,” IEEE Transactions on Communications, 48, 502, 2000.
[R6] E. I. Telatar and D. Tse, “Capacity and mutual information of broad-band multipath fading
channels,” in Proc. IEEE Int. Symp. Information Theory, 1998, p. 188.
[R7] R. Vaughan and J. B. Andersen, Channels, Propagation and Antennas for Mobile Communications.
London: The IEE, 2003.
[R8] P. Almers, E. Bonek, A. Burr, N. Czink, M. Debbah, V. Degli-esposti, H. Hofstetter, P. Ky¨osti, D.
Laurenson, G. Matz, A. F. Molisch, C. Oestges, and H. ¨Ozcelik, “Survey of channel and radio
propagation models for wireless MIMO systems,” EURASIP Journal on Wireless Communications
and Networking, 2007.
[R9] A. Paulraj, R. Nabar, and D. Gore, Introduction to Space-Time Wireless Communications. New
York: Cambridge University Press, 2003.
[R10] M. Steinbauer, A. F. Molisch, and E. Bonek, “The double-directional radio channel,” IEEE
Antennas and Propagation Magazine, vol. 43, no. 4, pp. 51 – 63, 2001.
[R11] M. Gustafsson and S. Nordebo, “Characterization of MIMO antennas using spherical vector
54
waves,” IEEE Transactions on Antennas and Propagation, vol. 54, pp. 2679 – 2682, Sep. 2006.
 References
[R12] A. A. Glazunov, M. Gustafsson, A. Molisch, F. Tufvesson, and G. Kristensson, “Spherical vector
wave expansion of Gaussian electromagnetic fields for antenna-channel interaction analysis,” IEEE
Transactions on Antennas and Propagation, vol. 57, no. 7, pp. 2055–2067, 2009.
[R13] B. Getu and J. Andersen, “The MIMO cube - a compact MIMO antenna,” IEEE Transactions on
Wireless Communications, vol. 4, pp. 1136 – 1141, May. 2005.
[R14] P. Almers, F. Tufvesson, and A. Molisch, “Keyhole effect in MIMO wireless channels:
measurements and theory,” IEEE Transactions on Wireless Communications, vol. 5, pp. 3596 – 3604,
Dec. 2006.
[R15] G. J. Foschini, “Layered space-time architecture for wireless communication in a fading
environment when using multi-element antennas”,” Bell Laboratories Technical Journal, pp. 41 – 59,
Autumn 1996.
[R16] P. Suvikunnas, J. Salo, L. Vuokko, J. Kivinen, K. Sulonen, and P. Vainikainen, “Comparison of
MIMO antenna configurations: Methods and experimental results,” IEEE Transactions on Vehicular
Technology, vol. 57, pp. 1021 – 1031, Mar. 2008.
[R17] H. ¨Ozcelik and C. Oestges, “Some remarkable properties of diagonally correlated MIMO
channels,” IEEE Transactions on Vehicular Technology, vol. 54, pp. 2143 – 2145, Nov. 2005.
[R18] D. S. Shiu, G. Foschini, M. Gans, and J. Kahn, “Fading correlation and its effect on the capacity of
multielement antenna systems,” IEEE Transactions on Communications, vol. 48, pp. 502 – 513, Mar.
2000.
[R19] M. S. Sharawi, “Printed Multi-Band MIMO Antenna Systems and Their Performance Metrics,”
IEEE Antennas and Propagation Magazine, Vol. 55, No. 5, pp. 218-232, October 2013.
[R20] S. Blanch, J. Romeu, and I. Corbella, “Exact representation of antenna system diversity
performance from input parameter description,” Electron. Lett., vol. 39, no. 9, pp. 705–707, May 2003.
55
 References
[R21] P. Hallbjorner, “The significance of radiation efficiencies when using S-parameters to calculate the
received signal correlation from two antennas,” IEEE Antennas Wireless Propag. Lett., vol. 4, pp. 97–99,
Jun. 2005.
[R22] S. H. Chae, S. K. Oh, and S. O. Park, “Analysis of Mutual Coupling, Correlations, and TARC in
WiBro MIMO Array Antenna,” IEEE Antennas and Wireless Propagation Letters, Vol. 6, pp. 122-125,
2007.
[R23] H. Shin and J. H. Lee, “Capacity of multiple-antenna fading channel: Spatial fading correlation,
double scattering, and keyhole,” IEEE Trans. Inf. Theory, vol. 49, pp. 2636–2647, Oct. 2003.

56
In wireless communications, channel state information (CSI) refers to known
channel properties of a communication link. This information describes how a
signal propagates from the transmitter to the receiver and represents the combined
effect of, for example, scattering, fading, and power decay with distance.

57