Reverberation Chambers:
Overview and Applications
Christopher L. Holloway
John Ladbury, Galen Koepke, Dave Hill, Kate Remley, William Young
National Institute of Standards and Technology
Electromagnetics Division
Boulder, Colorado
303-497-6184, email: holloway@boulder.nist.gov
OPEN AREA TEST SITES (OATS)
Problems:
Ambients
Reflections
Scanning
Interference
Positioning
TEM
Problems:
High Frequencies
Reflections
Test Volume
Positioning
GTEM
Problems:
Test Volume
Uniformity Along Cell
Positioning
ANECHOIC CHAMBER
Problems:
Low Frequencies
Reflections
Positioning
REVERBERATION CHAMBER
Why use reverberation chambers?
1) Relatively inexpensive
2) Relatively fast
The Classical OATS Measurements
The classic emissions test and standard limits (i.e, testing a
product above a ground plane at a specified antenna separation
and height) have their origins in interference problems with TV
reception.
Emissions Test Standard Problem
One problem with the emissions test standard is that it is based on an interference
paradigm (interference to terrestrial broadcast TV) that is, in general, no longer
valid nor realistic today. In a recent report, the FCC indicated that 85 % of
US households receive their TV service from either cable, direct broadcast
satellite (DBS), or other multichannel video programming distribution service,
and that only a small fraction of US households receive their TV via direct
terrestrial broadcast.
Coupling to TV antennas designed to receive terrestrial broadcast may no longer
be an issue.
EM/EMC Environment Today
•
In recent years, a proliferation of communication devices that are subject to
interference have been introduced into the marketplace.
•
Today, cell phones and pagers are used in confined offices containing personal
computers (PCs). Many different products containing microprocessors (e.g.,
TVs, VCRs, PCs, microwave ovens, cell phones, etc.) may be operating in the
same room.
•
Different electronic products may also be operating within metallic enclosures
(e.g., cars and airplanes). The walls, ceiling, and floor of an office, a room, a
car, or an airplane may or may not be highly conducting.
•
Hence, emissions from electric devices in these types of enclosures will likely
be quite different from emissions at an OATS. In fact, the environment may
more likely behave as either a reverberation chamber or a free space
environment.
Where Should We Test?
• Thus, would it not be better to perform tests more appropriate to
today's electromagnetic environment?
• Tests should be Shielded, Repeatable, Simple, Inexpensive, Fast,
Thorough, …
Commercial Solutions…
• Stirrer
• Turntable
Reverberation Chamber
Probe(s)
Receive Antenna
D-U-T
Transmit Antenna
Motor Control
Signal Generator
Amplifier
Directional coupler
Power Meter(s)
etc.
Control and Monitoring
Instrumentation for
Device-under-test
Receive Instrumentation:
Spectrum Analyzer
Receiver, Scopes
Probe System
etc.
Fields in a Metal Box (A Shielded Room)
Frozen Food
•In a metal box, the fields have well defined modal field distributions.
Locations in the chamber with very high field values
Locations in the chamber with very low field values
Fields in a Metal Box with Small Scatterer
In a metal box, the fields have well defined modal field distributions.
Small changes in locations where very high field values occur
Small changes in locations where very low field values occur
Fields in a Metal Box with Large Scatterer (Paddle)
Large changes in locations where very high field values occur
Large changes in locations where very low field values occur
In fact, after one fan rotation, all locations in the chamber will have the same
maxima and minima fields.
Stirring Method
TIME DOMAIN
Paddle
Click to play paddle
rotation
750MHz
Field Variations with Rotating Stirrer
Reverberation Chamber: All Shape and Sizes
Small Chamber
Large Chamber
Moving walls
NASA: Glenn Research Center (Sandusky, OH)
Reverberation Chamber with Moving Wall
Original Applications
• Radiated Immunity
 components
 large systems
• Antenna efficiency
• Calibrate rf probes
• RF/MW Spectrograph
• Radiated Emissions
 absorption properties
• Shielding
• Material heating
 cables
• Biological effects
 connectors
•
Conductivity and
 materials
material properties
Wireless Applications
•
Radiated power of mobile phones
•
Gain obtained by using diversity antennas in fading environments
•
Antenna efficiency measurements
•
Measurements on multiple-input multiple-output (MIMO) systems
•
Emulated channel testing in Rayleigh multipath environments
•
Emulated channel testing in Rician multipath environments
•
Measurements of receiver sensitivity of mobile terminals
•
Investigating biological effects of cell-phone base-station RF
exposure
Sampling Considerations
Techniques to Generate Samples
•
Mechanical techniques
 Paddle(s) or Tuner(s)
• stepped (tuned)
• continuous (stirred)
•
•
 Device-under-test and antenna position
 Moving walls (conductive fabric, etc.)
Electrical techniques (immunity tests)
 Frequency stirring
• stepped or swept
• random (noise modulation)
Hybrid techniques
Metric for a “Well” Performing Chamber
Standard bodies (e.g., IEC and CTIA) have various means of determining when a
chamber meets a performance metric. Typically we would like the chamber to be
“statistically uniform”.
DUT
Metric: The variance (or standard deviation) of the mean “stirred” power
(averaged over all paddle positions) for the twelve test points
are within some given value.
Metric for a “Well” Performing Chamber
NIST Chamber: 2.9 M by 4.2 m by 3.6 m
Metric for a Loaded Chamber
Using reverberation chambers for testing wireless devices is getting a lot of attention in recent years.
For the application, some type of rf absorber is used in the chamber in order to control the delay time
in the chamber.
Loading a chamber has an adverse effect on its “statistical uniformity”.
Loaded Chamber
EM / EMC / Wireless
Applications
HIDING Emission Problems
YOU CANNOT HIDE IN
Reverberation Chamber
Great for Emissions or Immunity Tests
In reverberation chambers you cannot hide emission problems.
Reverberation chambers will find problems.
EMISSION LIMITS
•Devices and/or products are tested for emissions to ensure that electromagnetic field strengths emitted
by the device and/or product are below a maximum specified electric (E) field strength over the
frequency range of 30 MHz to 1 GHz.
•These products are tested either on an open area test site (OATS) or in a semi-anechoic chamber.
•Products are tested for either Class A (commercial electronics) or Class B (consumer electronics) limits,
Class A equipment have protection limits at 10 m, and Class B equipment have protection limits at 3 m.
6.0E-4
5.0E-4
Class A: based on 10 m separation
| Emax | (V/ m)
4.0E-4
Class B: based on 3 m separation
3.0E-4
2.0E-4
1.0E-4
0.0E+0
0
200
400
600
Frequency (MHz)
800
1000
Total Radiated Power for Reverberation Chambers
Holloway et al., IEEE EMC Symposium
-40
Ptotal (dBm)
-45
-50
-55
Class A: based on 10 m separation
Class B: based on 3 m separation
-60
Class A: average limit
Class B: average limit
-65
0
100
200
300
400
500
600
Frequency (MHz)
700
800
900
1000
Emission Measurements
Relate total radiated power in reverberation
chambers to measurements made on OATS
with dipole correlation algorithms.
Spherical Dipole
-44
Reverb Chamber
OATS
-48
Reverb Chamber
-52
Anechoic Chamber
-52
-56
E-field [dBV/ m]
E-field (dBV/ m)
-48
-60
200
300
400
500
600
700
Frequency (Hz)
800
900
-56
1000
-60
-64
200
400
600
Frequency (Hz)
800
1000
Emissions Measurements of Devices
Great for Emissions or Immunity Tests
Total Radiated Power Comparison
Max Field Comparison
Shielding Properties of Materials
The conventional methods uses
normal incident plane-waves, i.e.,
Coaxial TEM fixtures.
E
i
In most applications, materials are exposed
to complex EM environments where fields are
incident on the material with various
polarizations and angles of incidence.
E
t
P 
SE = −10 Log10  t 
 Pi 
However, these approaches determine SE
for only a very limited set of incident
wave conditions.
Therefore, a test methodology that better
represents this type of environment would
be beneficial.
Nested Reverberation Chamber
Sample
Nested Reverberation Chamber
C.L. Holloway, D. Hill, J. Ladbury, G. Koepke, and R. Garzia, “Shielding
effectiveness measurements of materials in nested reverberation chambers,”
IEEE Trans. on Electromagnetic Compatibility, vol. 45, no. 2, pp. 350-356,
May, 2003.
 Pr ,in , s Pr ,o ,ns PrQ ,in ,ns Ptx ,in , s 

SE3 = −10 Log10 

P
P
P
P
r
in
ns
r
o
s
rQ
in
s
tx
in
ns
,
,
,
,
,
,
,
,


Sample
Nested Reverberation Chamber
80
Material 1
70
Material 2
Material 3
60
Material 4
90
Material 2: with coating
80
Material 2: with no coating
40
70
30
60
SE (dB)
a
20
50
10
90
40
a
0
80
1
30
10
Frequency (GHz)
70
Different Materials
20
60
10
SE (dB)
SE (dB)
50
1
50
10
Frequency (GHz)
Edge Effects
40
30
SE3: Chamber A
a
SE3: Chamber B
20
SE1: Chamber A
10
SE1: Chamber B
0
1
10
Frequency (GHz)
Different Chamber Sizes
Loading Effects: Rat in a Cage
Research Goal
Descritize in to FDTD
Grid
Take Real System
• To provide a way to gain
intuition into loaded
chamber responses to
better help measurements
• Ultimately correlate
measured and modeled
data, and use models to
predict what we cant
measure.
Simulate
• Develop a fast and efficient
numerical code to explore
multiple chamber
topologies. (FDTD)
Predict Stirred Fields Computationally to give us insight
into chamber measurements in a loaded environment
Measurements of Fake Rats
Lossy Body Configuration
Large Lossy Body
Distributed Lossy Bodies
NIH in the US is currently doing studies in 22 chambers
Periodic Lossy Bodies
Measuring Shielding for “small” Enclosures
reverb chamber
Pin
Pout
“small” enclosure
SE= -10 Log(Pin/ Pout): with frequency stirring
Problem with “small” Enclosures:
Measuring the Fields Inside
•However, Hill (IEEE Trans on EMC, 2005) has recently shown that the
statistics for the normal component of the E-field at the wall
are the same for the field in the center of the chamber.
1. Thus, small monopole (or loop) probes attached to the wall can be used to
determine the power in the center of a “small” enclosure.
2. Use frequency stirring for obtain samples.
Comparison with Different Reverberation
Chamber Approaches
port 3
port 2
port 2
port 1
port 1
port 4
Mode-Stirred with a Horn Antenna: SE => S31
Mode-Stirred with a Monopole Antenna: SE => S41
port 3
port 2
port 2
port 1
port 1
port 4
Frequency Stirring with a Horn Antenna: SE => S31
Frequency Stirring with a Monopole Antenna: SE => S41
Comparison with Different Approaches
20
20
18
mode_stirring_horn
16
freq_stirring_horn
14
18
mode_stirring_horn
16
freq_stirring_horn
14
mode_stirring_monopole
12
freq_stirring_monopole
SE (dB)
SE (dB)
mode_stirring_monopole
10
8
10
8
6
6
4
4
2
2
0
1000
2000
3000
4000
5000
6000
7000
8000
0
1000
9000 10000
freq_stirring_monopole
12
2000
3000
4000
Frequency (MHz)
20
7000
8000
9000 10000
20
18
mode_stirring_horn
18
16
freq_stirring_horn
16
mode_stirring_monopole
14
14
12
SE (dB)
SE (dB)
6000
(b) half-filled aperture
(a) open aperture
10
8
mode_stirring_horn
6
freq_stirring_horn
4
mode_stirring_monopole
freq_stirring_monopole
12
10
8
6
4
freq_stirring_monopole
2
0
1000
5000
Frequency (MHz)
2
2000
3000
4000
5000
6000
7000
8000
9000 10000
Frequency (MHz)
(c) narrow slot aperture
0
1000
2000
3000
4000
5000
6000
7000
8000
Frequency (MHz)
(d) generic aperture
9000 10000
Different Probe Lengths and Locations
30
50
Position 1, probe length=1.3 cm
Position 2, probe length=1.3 cm
45
25
Position 1, probe length=2.5 cm
40
a=0.95 cm
35
20
SE (dB)
30
SE (dB)
probe location 1
probe location 2
probe location 3
probe location 4
Position 2, probe length=2.5 cm
a=0.5 cm
25
15
20
10
15
10
5
5
0
3000
0
4000
5000
6000
7000
8000
Frequency (MHz)
9000
10000
11000
12000
2
2.5
3
3.5
4
Frequency (GHz)
This technique is currently being incorporated in the IEEE 299-1 standard
Antenna Measurement:
Reverberation Chambers are
Natural Multipath Environments
(Ideal for Antenna Measurements)
Radiation and Total Efficiencies of Antennas
Prad: radiated power
PA: available power
PR: reflected power
Pac: accepted power ===> Pac=PA-PR
•total efficiency: defined as the ratio of the power radiated to the power available
at the antenna port (including missmatch or reflection loss)
•radiation efficiency: defined as the ratio of the power radiated to the power
accepted by the antenna port
,
Problems with Current Two Approaches
The two current reverberation chamber approaches have the
following two problems:
• The first approach requires a reference antenna and it is assumed that the
efficiency of the reference antenna is known.
• The problem with the second approach is that we only have one expression
for the product of the two efficiencies. Thus, we “MUST” assume the two
antennas are identical in order to determine the efficiencies.
stirrer or paddle
stirrer or paddle
receive antenna ARx
PRx
PTx
Prf
antenna B
reference antenna Aref
antenna A
AUT
P2
P3
P1
VNA
ηAUT = (PAUT/Pref ) ηref
ηB ηA = [CRC / ω ].[<|S21|2> /τRC ]
Three Antenna Approach
Holloway et al., “Reverberation Chamber Techniques for Determining the Radiation and Total Efficiency of Antennas''
IEEE Trans. on Antenna and Propagation, vol. 60, no. 4, April 2012, pp. 1758-1770
stirrer or paddle
Antennas A & B
ηB ηA = [CRC / ω ] ΜΑΒ
ΜΑΒ =[<|S21|2> /τRC ]AB
antenna B
antenna A
antenna C
port 1
P3
port 2
Antennas A & C
P2
ηC ηA = [CRC / ω ] ΜΑC
P1
MAC =[<|S21|2> /τRC ]AC
VNA
Antennas B & C
ηC ηB = [CRC / ω ] ΜCB
Three equation and three unknown: ηA , ηB and ηC
MCB =[<|S21|2> /τRC ]AB
unknown
Three Antenna Approach
Total efficiency (mismatch and ohmic loss)
Radiation efficiency (mismatch correction)
One Antenna Approach
Holloway et al., “Reverberation Chamber Techniques for Determining the Radiation and Total Efficiency of Antennas''
IEEE Trans. on Antenna and Propagation, vol. 60, no. 4, April 2012, pp. 1758-1770
PTX=ηA P1
;
Prf=P2 / ηA
<Prf>/ PTX = <S11> / (ηA * ηA )
For an “ideal” chamber it can be shown that: <Prf>=2 <PRX>.
This result is analogous to the enhanced backscatter that has
been derived for scattering by a random medium.
(ηA )2= (<|S21|2> /2) * (<PRX>/ PTX) and <PRX>/ PTX = Q / CRC
known
ηA =
unknown
C RC
2ω
S11
τ RC
2
Two Antenna Approach
stirrer or paddle
Following a similar procedure as the
one-antenna method, we can show:
PRx
PTx
Prf
antenna B
S11
antenna A
2
= η Atotalη Atotal
Prf ,1
= η Btotalη Btotal
Prf ,2
= η Btotalη Atotal
PRX
P2
P3
S 22
P1
VNA
S 21
2
2
PTX
PTX
PTX
=
eb PRX
b PRX and Prf ,2
If we assume
that Prf ,1 e=
, where eb is the
unknown
enhanced backscatter constant (and is 2 for an “ideal” chamber),
then the two-antenna method reduces to the following:.
η Atotal =
η
total
B
S11
2
CRC
ω eb τ RC
CRC
=
ω eb
S 22
τ RC
where eb =
2
S 22
2
S 21
S 22
2
2
Experimental Data for Three Antennas
Antenna A: 13.5 cm by 22.5 cm horn
Antenna B: 13.5 cm by 24.5 cm horn
Antenna C: 9 cm monopole on 45.5 cm ground plane
Experimental Data for Three Antennas
We performed three sets of measurements with a VNA:
1)
2)
3)
Antennas A & B
Antennas A & C
Antennas B & C
Determining S11
.
stirrer or paddle
PTx
Prf
antenna A
P3
P1
VNA
Determining τRC
We determine τRC from time-domain data
Power Delay Profile:
PDP(τ ) = h(t )
2
where h(t) is the Fourier transform of S21(ω)
slope=1/τRC
Determining τRC
2200
2100
2000
1900
Ant A: S11 (A to B)
(ns)
1700
τRC
1800
1600
Ant A: S11 (A to C)
Ant A: S21 (A to B)
Ant A: S21 (A to C)
Ant B: S11 (B to A)
Ant B: S11 (B to C)
1500
Ant B: S21 (B to A)
1400
Ant B: S21 (B to C)
1300
Ant C: S11 (C to B)
Ant C: S11 (C to A)
Ant C: S21 (C to A)
1200
Ant C: S21 (C to B)
1100
1
2
3
4
Frequency (GHz)
5
6
This shows that τRC is independent of location in the chamber and
independent of the antenna use in the measurement.
Radiation Efficiency
1
0.98
0.96
0.94
η rad
0.92
0.9
0.88
η rad
A
0.86
η rad
B
0.84
η rad
C
0.82
manufacturer
computed data
0.8
1
2
3
4
Frequency (GHz)
5
6
The manufacturer of antenna B state the radiation efficiency is 91 %.
(Danny Odum and Dr. Vince Rodriguez of ETS-Lindgren)
We see that the monopole antenna has the least ohmic loss of the three antennas.
Antenna B has the largest ohmic losses.
Total Efficiency
1
0.9
0.8
0.7
η total
0.6
0.5
0.4
η total
A
0.3
0.2
η total
B
0.1
η total
C
0
1
2
3
4
Frequency (GHz)
5
6
We see that the monopole antenna has the lowest total efficiency of the three
antennas. This obviously implies that the monopole is the worst impedance matched
antenna of the three.
Measured S11 for the Three Antennas
Antenna A
Antenna C
0
0
-5
-5
-10
-10
-15
-20
|S11| (dB)
Antenna B
-25
0
-30
Brass Horn
Reverberation Chamber
Anechoic Chamber
-35
-5
-40
-10
-45
-15
-50
-20
1000
2000
3000
4000
Frequency (MHz)
5000
6000
|S11| (dB)
|S11| (dB)
-15
-20
-25
-30
-35
Monopole
Reverberation Chamber
Anechoic Chamber
-40
-45
-50
-25
1000
-30
2000
3000
4000
Frequency (MHz)
Silver Horn
Reverberation Chamber
Anechoic Chamber
-35
-40
-45
-50
1000
2000
3000
4000
Frequency (MHz)
5000
6000
These comparisons illustrate that RC and anechoic chamber measurements
give the same result for S11
5000
6000
Comparisons to Other Chambers and Techniques
Microstrip
Log-periodic
1
ITT RC
Anechoic chamber
Numerical calculations
0.9
1
0.9
0.7
0.8
0.6
0.7
0.5
0.6
ηtotal
0.8
0.4
0.5
NIST RC
ITT RC
Anechoic chamber
0.4
0.3
0.3
0.2
Horn
0.1
0.2
1
0.1
0
1480
1520
1560
1600
1640
Frequency (MHz)
1680
0
1720 0.9
0
0.8
0.7
ηtotal
ηtotal
1.1
0.6
NIST RC
ITT RC
Anechoic chamber
0.5
0.4
0.3
0.2
1
2
3
4
Frequency (GHz)
5
6
400
800
1200
Frequency (MHz)
1600
2000
Measurement Uncertainties
Uncertainty Source
uncertainty
Type A:
0.45 dB
0.048 dB
uS21
uτRC
Total: Type A
Type B:
uVNA
Total: Type B
Total:
0.45 dB
0.2 dB
0.2 dB
0.49 dB
Total uncertainties for the three-antenna approach is
0.49 dB or 12 %
However, by increasing the number of independent samples and employing a
combination of paddle-averaging, frequency-averaging, and positionaveraging (a common practice in RC measurements), the uncertainties can be
reduced further to below 10 %.
Magnetic Antenna
Comparison to Horn
New Magnetic Antenna
Loop Antenna
Total Radiated Power Relative to Horn (dB)
5
0
-5
-10
-15
-20
-25
-30
-35
-40
-45
-50
200
Magnetic EZ-Antenna: Ground Plane-Center of Chamber
Magnetic EZ-Antenna: No Ground Plane-Center of Chamber
Magnetic EZ-Antenna: On Chamber Wall
3 cm Loop antenna
250
300
Frequency (MHz)
350
400
Comparison of Total Efficiency with Other Methods
Holloway et al., “Reverberation Chamber Techniques for Determining the Radiation and Total Efficiency of Antennas''
IEEE Trans. on Antenna and Propagation, vol. 60, no. 4, April 2012, pp. 1758-1770
Antenna B
Antenna A
1
1
0.95
0.95
0.9
0.9
0.85
0.85
0.8
η total
B
0.75
1
η total,
A
0.7
Antenna C
2 : (A and B)
η total,
A
0.65
2 : (A and C)
η total,
A
0.6
3
4
Frequency (GHz)
5
0.65
2 : (B and C)
η total,
B
3 : (A, B, and C)
η total,
B
0.5
0.8
2
2 : (B and A)
η total,
B
0.9
1
0.5
1
1
η total,
B
0.7
0.55
3 : (A, B, and C)
η total,
A
0.55
0.75
0.6
1
0.7
6
0.6
η total
C
η total
A
0.8
0.5
0.4
1
η total,
C
2 : (C and A)
η total,
C
0.3
2 : (C and B)
η total,
C
0.2
3 : (A, B, and C)
η total,
C
0.1
0
1
2
3
4
Frequency (GHz)
5
6
2
3
4
Frequency (GHz)
5
6
Standardization of Wireless Measurements
Can we use a reverberation chambers as a reliable and
repeatable test facilities that has the capability of simulating
any multipath environment for the testing of wireless
communications devices?
If so, such a test facility will be useful in wireless measurement
standards.
Total Radiated Power from Cell-Phones
Data from CTIA working group on total radiate power (TRP) testing:
TRP Comparison, Free Space, Slider Open, W-CDMA Band II
5.00
4.50
Anechoic Lab A Band II
Anechoic Lab B Band II
Anechoic Lab C Band II
Reverb Lab A Band II
Reverb Lab B Band II
Reverb Lab C Band II
Reverb Lab D Band II
Reverb Lab E Band II
Relative TRP
4.00
3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
Low
Mid
High
Reference Channel
TRP Comparison, Free Space, Slider Open GSM850
5.00
4.50
Anechoic Lab A GSM850
Anechoic Lab B GSM850
Anechoic Lab C GSM850
Reverb Lab A GSM850
Reverb Lab B GSM850
Reverb Lab C GSM850
Reverb Lab D GSM850
Reverb Lab E GSM850
Relative TRP
4.00
3.50
3.00
2.50
2.00
1.50
1.00
Reverb chamber data has less variability than the anechoic data!
0.50
0.00
Low
Mid
Channel
High
Multipath Environments
Multipath Environments
Extensive measurements have shown that when light of sight (LOS) path is present
the radio multipath environment is well approximated by a Ricean channel, and
when no LOS is present the channel is well approximated by a Rayleigh channel:
N
E = ALOS cos(2π f c t ) + ∑ An cos[2π ( f c + f n )t + φn ]
n =1
The Amplitude of E is either Rayleigh or Ricean depending if a LOS path is present.
Urban Environment
Rural Environment
Ricean K-factor
K-factor:
k=
direct component
scattered components
or
K = 10 Log ( k )
K= - ∞ dB
(Rayleigh)
K=1 dB
K=4 dB
K=10 dB
Reverberation Chambers are
Natural Multipath Environments
Typical Reverberation Chamber Set-up
Antenna pointing away from probe (DUT)
a
metallic walls
Paddle
DUT
Transmitting Antenna
a
A Rayleigh test environment
Can we generate a Ricean environment?
Chamber Set-up for Ricean Environment
Antenna pointing toward (DUT)
metallic walls
Paddle
DUT
Transmitting Antenna
We will show that by varying the characteristics of the reverberation
chamber and/or the antenna configurations in the chamber, any desired
Rician K-factor can be obtained.
Reverberation Chamber Ricean Environment
It can be shown: see Holloway et al, IEEE Trans on Antenna and Propag., vol. 54, no. 11, pp.
3167-3177, November 2006.
K=
3 V D
2 λ Q r2
Note:
•We see that K is proportional to D. This suggests that if an antenna with a well defined antenna
pattern is used, it can be rotated with respect to the DUT, thereby changing the K-factor.
•Secondly, we see that if r is large, K is small (approaching a Rayleigh environment);
if r is small, K is large. This suggests that if the separation distance between the antenna
and the DUT is varied, then the K-factor can also be changed to some desired value.
•Next we see that by varying Q (the chamber quality factor), the K-factor can be changed to some
desired value. The Q of the chamber can easily be varied by simply loading the chamber
with lossy materials.
Also, if K becomes small, the distribution approaches Rayleigh.
Thus, varying all these different quantities in a judicious manner can result in
controllable K-factor over a reasonably large range.
Measured K-factor for Different
Antenna Separation
1.0E+02
0.5 m
1 m separation
1.0E+01
2 m separation
K-factor
1.0E+00
1.0E-01
1.0E-02
1.0E-03
1.0E-04
1000
2000
3000
4000
5000
6000
7000
Frequency (MHz)
Each set of curves represents a different distance of separation. The thick black curve
running over each data set represents the K-factor obtained by using d determined in
the anechoic chamber.
Measured K-factor for
Chamber Loading
1.0E+03
1.0E+02
6 pcs absorber
2 pcs absorber
K-factor
0 pcs absorber
1.0E+01
1.0E+00
1.0E-01
1.0E-02
1000
2000
3000
4000
5000
6000
7000
Frequency (MHz)
The thick black curve running over each data set represents the K-factor obtained
by using d determined in the anechoic chamber.
Measured K-factor for
Different Antenna Rotations
1.0E+02
0 degrees
30 degrees
K-factor
1.0E+01
1.0E+00
90 degrees
1.0E-01
1.0E-02
1.0E-03
1000
2000
3000
4000
5000
6000
7000
Frequeency (MHz)
The thick black curve running over each data set represents the K-factor obtained
by using d determined in the anechoic chamber. Each data set was taken at 1 m separation
and with 4 pieces of absorber in the chamber.
Measured K-factor for
Different Antenna Polarizations
1.0E+03
co-polarized
1.0E+02
45 degree polarization
K-factor
1.0E+01
1.0E+00
cross-polarized
1.0E-01
1.0E-02
1.0E-03
1.0E-04
1000
2000
3000
4000
5000
6000
7000
Frequency (MHz)
The thick black curve running over each data set represents the K-factor obtained
by using d determined in the anechoic chamber. Each data set was taken at 1 m separation
and with 4 pieces of absorber in the chamber.
Simulating Propagation Environments
with Different Impulse Responses and rms Delay Spreads
-13
x 10
-12
10
x 10
7
6
PDP (linear)
PDP (linear)
8
6
5
4
Can we simulate these different PDP(t) in a reverberation chamber?
Transmitting
antenna
Receiving
antenna
4
3
2
Rx8
τrms = 105 ns
2
tripods
62 inches
VNA
Port 1
Ground Plane
Fiber Optic
Transmitter
RF Optical
200 m
Optical
Fiber
Fiber Optic
Receiver
RF Optical
Port 4
0
1000
2000
Delay (ns)
3000
Rx12
τrms = 232 ns
1
0
1000
2000
Delay (ns)
3000
S21 Measurements: Loading the Chamber
Impulse Responses and Power Delay Profiles
Power Delay Profile:
PDP(τ ) = h(t )
2
where h(t) is the Fourier transform of S21(ω)
Loading the Chamber
rms Delay Spreads
One characteristic of the PDP that has been shown to be particularly important
in wireless systems that use digital modulation is the rms delay spread of the PDP:
∞
τ rms =
∫ (t − τ
o)
2
P(t ) dt
0
∞
∫ P(t ) dt
0
where τo is the mean delay of the propagation channel, given by
∞
τo =
∫ tP(t ) dt
0
∞
∫ P(t ) dt
0
Impulse Responses and rms Delay Spreads
(200 MHz bandpass filter on S21 data)
Instantaneous Results Can Vary
RMS Delay Spread at Each Paddle Position
200
All PDPs, Antenna Position 2a, Abs Ht 2, Xpol
20
180
RMS Delay Spread (ns)
140
120
100
80
3 Abs Floor, Horn Pos 1
3 Abs Floor, Horn Pos 2
1 Abs Floor, 2 Abs Raised, Horn Pos 1
1 Abs Floor, 2 Abs Raised, Horn Pos 2
3 Abs Raised, Horn Pos 1
3 Abs Raised, Horn Pos 2
60
40
20
0
0
20
40
60
Paddle Position
80
100
Power Delay Profile (linear units)
18
160
16
14
12
10
8
6
4
2
0
0
100
200
300
400
500
Delay (ns)
600
700
800
rms Delay Spreads from Q measurements
τ rms =
Q
ω
2α ln(α ) − ln 2 (α )ε − 2α + 2 (α ln(α ) + 1 − α ) 2
−
(1 − α ) + K
((1 − α ) + K ) 2
where K is the K-factor and α threshold.
Thus, once we have Q, we can estimate τrms
Impulse Responses and rms Delay Spreads
for Different Ricean K-factors
Testing Wireless Devices in
Realistic Environments
Difficult Radio Environments
Office Corridor
Oil Refinery
Apartment Building
Subterranean Tunnels
NIST is measuring signal penetration and multipath in representative emergency response
environments to provide data for improved wireless device design, standards development,
and better channel models.
How Well Can we Simulate a Real Envirnoment?
Power Delay Profile in an oil refinery.
Example: Denver Highrise Tests
18
21
10
15
14
13
12
11
8
5
7
6
3
19
20
17
16
9
4
North
1
West
2
East
Transmitting
antenna
Receiving
antenna
South
VNA measurement test locations are in pink
tripods
62 inches
VNA
Port 1
Ground Plane
Fiber Optic
Transmitter
RF Optical
200 m
Optical
Fiber
Fiber Optic
Receiver
RF Optical
Port 4
Replicate Environment in Reverb Chamber
horn
antenna
mode stirrers
wireless
device
antenna
absorber
0
1 absorber: τrms=187 ns
-5
-10
3 absorber: τrms=106 ns
Power Delay Profile (dB)
-15
7 absorber: τrms=66 ns
-20
-25
-30
-35
-40
-45
-50
-55
-60
Reverberation chamber with absorbing material
and phantom head
Large office biulding
τrms=59 ns
-65
-70
0
50
100
150
200
250
300
time (ns)
350
400
450
500
Adding More Realism to the PDP: Urban Canyon
-90
-100
-110
Noise Threshold = -116 dB
-120
rms
-110
-120
PDP (dB)
PDP (dB)
TX1 to RX5
τ
= 115 ns
-130
-140
-130
-140
-150
-150
-160
-170
0
200
400
600
800
Delay (ns)
1000
1200
-160
TX1 to RX9
τ
= 39 ns
-170
Noise Threshold = -116 dB
0
rms
500
Delay (ns)
R9
1000
R10
R5
•Mean of 27 NLOS
measurements made in Denver
urban canyon.
•Channel characterization and
wireless device measurements.
T1
T
R1212
T3
T2
T
Combine Fading Simulator with Reverb
Chamber
•Fading simulator
replicates delayed,
scaled versions
•Reverb chamber
introduces exponential
profile
Pulse out
•Pulse generator used to
amplitude modulate RF,
creates short-duration
pulse
Wide AM in
Pulse Generator
Trigger out
Vector Signal
Generator
RF out
Fading System
Reverberation
Chamber
Fading Simulator
Input
Output
Amp
Stirrer
Ch. 1 input
(Trigger)
Ch. 4
input
GPIB
Real Time
Oscilloscope
Clustering of Multipath is Common in
Urban Environments
Shortest path
Measured Denver
fitted simulated data
Measured
reverb
measured Data
Denver
Analytic curve fit
1
0.9
0.8
TX
0.7
PDP
0.6
0.5
0.4
0.3
RX
0.2
Welton Street
0.1
RX
1
0
100
200
300
400
500
600
700
Delay [ns]
Excite reverberation
chamber with channel
emulator to create
multipath clusters
800
900
RX
2
RX
3
Clusters of
exponential
distributions off of
buildings
1000
Parking lot
17 th Street
0
TX
Transmitter Site
Glenarm Pl
NIST’s Goal: Lab-Based Test Methods
Location
and Notes
Republic
Notes:
- System 1
repeater at
test point 2.
Test
Point
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
VNA Loss
Data (dB)
7.23
27.06
38.15
37.60
37.18
42.26
46.04
44.88
48.30
45.34
50.25
50.48
50.98
51.82
49.60
44.64
29.28
30.45
42.24
39.30
47.07
Path Loss
@700 MHz
(dB)
68.6149
88.4449
99.5349
98.9849
98.5649
103.6449
107.4249
106.2649
109.6849
106.7249
111.6349
111.8649
112.3649
113.2049
110.9849
106.0249
90.6649
91.8349
103.6249
100.6849
108.4549
RMS Delay
Spread
@700 MHz
(ns)
44.99
39.52
52.30
133.41
81.25
102.78
138.29
104.69
376.10
338.17
167.91
231.57
209.07
192.25
240.20
377.45
296.87
161.75
429.90
333.25
453.47
18
21
10
15
14
13
12
11
8
5
7
6
3
19
20
9
4
North
1
West
2
East
South
VNA measurement test locations are in pink
Replicate field-tested wireless device performance in the
reverberation chamber
Used for standardized testing of wireless devices:
• public-safety community (NFPA)
• wireless sector (CTIA)
17
16
BER Measurements - setup
Agilent 4438C Vector Signal Generator
Reverberation chamber
External trigger
Agilent 89600 Vector Signal Analyzer
GPIB connection
to control VSG
Firewire connection
to control VSA
BER Measurements
BER for a 243 ksps BPSK signal
BER for a 786 ksps BPSK signal
How Well Can we Simulate a Real Envirnoment?
BER measurement in a laboratory.
Reverberation Chamber Test
Environment for MIMO Systems
Motivation
SISO
Alternative transmit/receive
configurations can improve wireless
reception in weak-signal and multipath
environments
To verify performance of multiple
antenna algorithms, testing in a
multipath environment is desirable
We discuss methods for implementing
such test environments using
reverberation chambers
RX
TX
MISO
TX
RX
SIMO
TX
RX
MIMO
TX
RX
Measurement Set-up
•Multiple TX simulated using 2
VSGs
•Vector signal analyzer provides
channel power and demodulated
data
•BPSK modulated signal;
random, equal distribution; 2048
bits
LO in
LO out
I out
Q out
Event 1
GPIB
VSG 1
RF out
LO in
LO out
I in
Q in
Event 1
GPIB
IEEE 1394
Firewire
VSG 2
Pattern sync.
VSA
RF out
•Error correction only to recover
constellation after deep fade
LO in
•Paddle stepped
Reverberation chamber
Stirrer
RF in
NIST Chamber Characteristics
•Chamber dimensions:
4.28m x 3.66m x 2.90m
•Two paddles: vertical
and horizontal
•Table shows Q, RMS
delay spread, and
coherence bandwidth
for various numbers of
absorbing blocks
Absorber
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Q
47130
8505
3319
2051
1479
1407
1067
918
765
731
550
551
435
399
369
340
332
τ RMS [ns]
3121.99
563.38
220.08
136.01
98.06
93.32
70.66
60.82
50.71
48.48
36.49
36.55
28.83
26.47
24.48
22.57
22.03
Δf [MHz]
0.32
1.77
4.54
7.35
10.20
10.71
14.14
16.42
19.72
20.62
27.40
27.36
34.68
37.77
40.84
44.30
45.38
SIMO
Real life
Simulation in reverberation chamber
A
Stirrer
B
A
Reverberation
chamber
Transmitter
B
Receiver
Antenna A receives strongest
signal
Vector
Signal
Generator
•Also called Diversity
•Power meter monitors received
signal strength
•Strongest signal demodulated
by receiver
Vector
Signal
Analyzer
Data A
Data B
Switch
RX Power A
RX Power B
RX Power A is strongest
In post processing Data A is selected
•VSA provides channel power
•Data from strongest signal
chosen in post processing
MIMO
Real life
Simulation in reverberation chamber
Stirrer
A
C
Reverberation
chamber
B
Transmitter
D
Receiver
Various schemes:
•Precoding (like beamforming)
•Spatial multiplexing (data
broken into multiple streams, TX
on uncorrelated channels)
•Diversity coding (identical data
streams TX using orthogonal
codes. RX decides which one to
use)
Vector
Signal
Analyzer
Vector
Signal
Generator 1
Vector
Signal
Generator 2
Data A-C
Data A-D
Data B-C
Data B-D
Switch
RX Power A-C
RX Power A-D
RX Power B-C
RX Power B-D
•MISO => MIMO using post
processing
Measured Results:
Various Data Rates
30
25
BER [%]
20
SISO
15
Receive diversity (MISO)
10
•Best performance:
RX diversity with TX beamforming
and MIMO
5
•Huge increase in BER for high data
rates may be affected by coherence
BW of chamber
30
0
768 1 Msps 2.5 3 Msps 5 Msps 10
ksps
Msps
Msps
25
20
BER [%]
Frequency = 2.4 GHz
Pout = -50 dBm
BPSK modulated signal
MIMO without
beamforming
TX beamforming (SIMO)
15
Receive diversity with TX
beamforming
10
MIMO
5
0
768 1 Msps 2.5 3 Msps 5 Msps 10
ksps
Msps
Msps
Number of Absorbers
5
4,5
BER SISO
4
BER MISO
3,5
BER [%]
BER SIMO
3
BER MIMO no Beamforming
2,5
BER MIMO
2
1,5
1
0,5
Number of absorbers inside chamber
Frequency = 2.4 GHz
Pout = -50 dBm
BPSK modulated signal, 768 ksps
70%
12
11
10
9
8
7
6
5
4
3
2
1
0
Modulated-Signal, BER Measurements
BER as a function of
received power: BER
increases for
• lower power levels
• wide modulation
bandwidths
Coherence bandwidth of
chamber interacts with
signal bandwidth
MIMO Antenna Testing
CTIA “reference antennas” developed to ensure test chamber’s ability
to distinguish “good”, “nominal”, and “bad antenna performance
NIST Reverberation Chamber
Ref. Antenna
(Ports 1 & 2)
RF chokes
1
Monopole
(Port 3)
CTIA Antennas
Monopole
(Port 4)
Measurement Set Up
2
3
4
VNA
MIMO Antenna Correlation and
Capacity
No power/efficiency
normalization
Correlation Coefficient Magnitudes
for Different Positions/Orientations
Median capacity at 50th
Percentile (bps/Hz):
{ 3.5, 4.6, 5.6 }
• Capacity: Provides upper bound on throughput
• Distinction between antennas due to inclusion of antenna efficiencies.
TRP: Machine-to-Machine Applications
M2M: Automated wireless
• ATMs, parking meters, home
security systems, etc.
• Some are WiFi, GPS enabled:
• Unlicensed: open to
interference
• Multiple radios/antennas
• Coexistence
OTA Measurement Challenges
• Device dimensions often a
significant fraction of working
volume of test chamber
• Calibrations affected by antenna
and device placement
Goal of NIST work
• Guidelines for use of
reverberation chambers for largeform-factor device testing
GPS
Satellite
Remote
Server
Parking
Meter
TRP: M2M Experimental Set-up
Large form factor devices will load the chamber, affect calibrations
M3
M2
M11
M9
M10
Horn
P3
P1
Boxes/Absorbers
P2
M12
M6
M7
Monopoles
2 and 3
M8
M1
M4
TX
Antenna
M5
Experiment:
• 12 omnidirectional receive antennas tuned to 1.9 GHz
• Absorbing and reflective objects of various sizes
M2M: Metallic and Absorbing Objects
Absorber: damps out reflections when compared to metallic boxes
Metallic boxes: minor but measurable loading effects
Power Delay Profile (Monopole 1)
-50
-60
Metal
Boxes
0
775
4
708
8
650
-100
10
623
-110
12
590
-70
-80
dB
Loading τrms
(ns)
zero boxes
four boxes
eight boxes
ten boxes
twelve boxes
2 RF Abs
-90
Absorber
-120
-130
0
2
4
6
Time (µs)
2 RF Abs 192
8
10
12
• Calibration and measurement guidelines must consider largeform-factor, absorbing devices
• Multiple-antenna M2M:
Realistic MIMO Environments
Non-line-of-sight: Nested Reverberation Chamber
Replicate room-to-room or
indoor-to-outdoor types of
environments
Line of sight, with
nested chamber
Line of sight,
no nested chamber
1. The paddle in the outer
chamber is turned through
100 positions, over 360⁰.
2. Paddle in the nested chamber
is turned continuously.
No line of sight,
with nested chamber
Measuring MIMO Channel Correlation using a Nested
Reverberation Chamber
Large, outer
Chamber
S42,S24
S42,S41
S41,S31
S42,S32
S41,S32
S31,S32
1
MIMO Antenna
(Ports 1 & 2)
Chamber Configurations (2.4 GHz):
1) No stirring in small chamber
2) Stirring in both; no RF absorbers
3) Stirring in both, one RF absorber
4) Stirring in both, two RF
absorbers
Nested Chamber
|σ S#,S#|
0.8
0.6
Low correlation in
S-parameters =
good MIMO channel
0.4
0.2
MIMO Antenna
(Ports 3 & 4)
0
1
2
3
4
Chamber Configuration
Nested Chamber Wireless Environment
300
LOS, w/o nested chamber
LOS, w/ nested chamber, w/ small paddle
NLOS, w/nested chamber, w/o small paddle
NLOS, w/ nested chamber, w/ small paddle
RMS Delay Spread (ns)
250
200
150
100
50
0
0
10
20
30
40
Threshold (dB)
50
60
70
“Tune” power delay profile (RMS delay spread) with
• nested chamber orientation to TX antenna
• various stirring algorithms
80
Reverberation Chambers for Wireless
Device Testing
Reverberation chambers represent reliable and repeatable test
facilities that have the capability of simulating multipath
environments for the testing of wireless communications devices.
Ongoing research:
•
•
•
•
•
•
Direct path with omnidirectional antennas
Tuning decay times = field non-uniformity
How to test devices with repeaters
Creating complicated PDPs for wireless device test
Automating PDP development
Advanced transmission, multiple antenna systems
• Test methods (CTIA, 3GPP groups)
• Angle of arrival measurements
• Uncertainties
Reverberation Chamber Standards Proposed
Testing Methods
Standards
•International Standard IEC 61000-4-21: Testing and measurement
techniques – Reverberation chamber test methods
•IEEE 299.1: Testing shielded enclosures
•3rd Generation Partnership Project (3GPP) RAN4
– R4-111690, “TP for 37.976: LTE MIMO OTA Test Plan for
Reverberation Chamber Based Methodologies”
Testing Methods
•CTIA Certification Program Working Group Contribution
– RCSG090101, P.-S. Kildal and C. Orlenius, “TRP and TIS/AFS
Measurements of Mobile Stations in Reverberation Chambers (RC)”
• “Utilizing a channel emulator with a reverberation chamber to create the
optimal MIMO OTA test methodology”
– C. Wright, S. Basuki, [8]
Summary
• Reverberation chamber measurements are
thorough and robust.
• Proper sampling techniques reduce
measurement uncertainties
• Statistical models help minimize the
number of samples required
Summary
• Reverberation chambers capture radiated.
• Results are insensitive to EUT placement in
the chamber
• Results are independent of EUT or antenna
radiation pattern
• Enclosed system free from external
interference
• Relatively inexpensive
• Relatively fast
References from NIST on Wireless Measurements
in Reverberation Chambers
[1] C.L. Holloway, D.A. Hill, J.M. Ladbury, P. Wilson, G. Koepke, and J. Coder, “On the Use of Reverberation
Chambers to Simulate a Controllable Rician Radio Environment for the Testing of Wireless
Devices”, IEEE Transactions on Antennas and Propagation, Special Issue on Wireless
Communications, vol. 54, no. 11, pp. 3167-3177, Nov., 2006.
[2] E. Genender, C.L. Holloway, K.A. Remley, J.M. Ladbury, G. Koepke, and H, “Simulating the Multipath
Channel with a Reverberation Chamber: Application to Bit Error Rate Measurements,” IEEE
Transactions on EMC, vol. 52, no 4, pp. 766 – 777, Nov. 2010.
[3] E. Genender, C.L. Holloway, K.A. Remley, J. Ladbury, G. Koepke and H. Garbe, “Using Reverberation
Chamber to Simulate the Power Delay Profile of a Wireless Environment”, EMC Europe 2008,
Sept, 2008, Hamburg, Germany.
[4] H. Fielitz, K.A. Remley, C.L. Holloway, Q. Zhang, Q. Wu, and D. W. Matolak, “Reverberation-Chamber
Test Environment for Outdoor Urban Wireless Propagation Studies”, IEEE Antennas and Wireless
Propag. Lett., 2009.
[5] K.A. Remley, H. Fielitz, and C.L. Holloway, Q. Zhang, Q. Wu, and D. W. Matolak, “Simulation of a MIMO
system in a reverberation chamber”, IEEE EMC Symp. August 2011
[6] K.A. Remley, S.J. Floris, and C.L. Holloway, “Static and Dynamic Propagation-Channel Impairments in
Reverberation Chambers,” submitted to IEEE Transactions on EMC, 2010.
[7] D. Hill, “Electromagnetic Fields in Cavities: Deterministic and Statistical Theories” , IEEE Press, Copyright
© 2009.
Other References on Wireless Measurements
in Reverberation Chambers
[8] C. Wright, S. Basuki, "Utilizing a channel emulator with a reverberation chamber to create the optimal
MIMO OTA test methodology," Mobile Congress (GMC), 2010 Global , vol., no., pp.1-5, 18-19
Oct. 2010.
[9] N. Serafimov, P.-S. Kildal, and T. Bolin, “Comparison between radiation efficiencies of phone antennas and
radiated power of mobile phones measured in anechoic chambers and reverberation chambers,”
in Proc. IEEE Antennas Propag. Int. Symp. 2002, Jun. 2002, vol. 2, pp.478–481.
[10] P.-S. Kildal, K. Rosengren, J. Byun, and J. Lee, “Definition of effective diversity gain and how to measure
it in a reverberation chamber,” Microwave Opt. Technol. Lett., vol. 34, no. 1, pp. 56–59, Jul. 2002.
[11] K. Rosengren and P.-S. Kildal, “Radiation efficiency, correlation, diversity gain, and capacity of a six
monopole antenna array for a MIMO system: Theory, simulation and measurement in reverberation
chamber,” Proc. Inst. Elect. Eng. Microwave, Antennas, Propag., vol. 152, no. 1, pp. 7–16, Feb.
2005.
[12] M. Lienard and P. Degauque, “Simulation of dual array multipath channels using mode-stirred
reverberation chambers,” Electron. Lett., vol. 40, no. 10, pp. 578–5790, May 2004.
[13] P.-S. Kildal and K. Rosengren, “Electromagnetic analysis of effective and apparent diversity gain of two
parallel dipoles,” IEEE Antennas Wireless Propag. Lett., vol. 2, pp. 9–13, 2003.