MIMO and other Wireless Measurements in
Reverberation Chambers at NIST
Presented by: William Young
C. L. Holloway, K.A. Remley, J. Ladbury, G. Koepke, W.F. Young,
R. Pirkl, E. Genender, S. Floris, H. Fielitz
National Institute of Standards and Technology
Electromagnetics Division
Boulder, Colorado
303-497-3471, email: wfy@boulder.nist.gov
July 2011, SPOKANE, WA
Content
• Basics of Radio Frequency Reverberation
Chambers
• Key Reverberation Chambers Characteristics for
Wireless Measurements
• Creating Realistic Environments for Testing
Wireless Devices
• Reverberation Chamber Test Environments for
MIMO Systems
1
BASICS OF RADIO
FREQUENCY
REVERBERATION CHAMBERS
OPEN AREA TEST SITES (OATS)
Problems:
Ambients
Reflections
Scanning
Interference
Positioning
2
ANECHOIC CHAMBER
Problems:
Low Frequencies
Reflections
Positioning
Directional Antennas
REVERBERATION CHAMBER
Why use a reverberation chamber?
1) Relatively inexpensive
2) Relatively fast
3) Multipath environment- useful
for MIMO testing
3
Commercial Solutions…
• Stirrer
• Turntable
Reverberation chambers are being used for a
wide range of EM and EMC measurements.
Reverberation Chamber
Probe(s)
Transmit Antenna
Receive Antenna
D-U-T
Motor Control
Transmit Instrumentation:
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.
4
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
(Paddle)
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
5
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 nearly the
same maxima and minima fields.
Stirring Method
TIME DOMAIN
Paddle
Click to play paddle
rotation
750MHz
6
Reverberation Chamber: All Shape and Sizes
Small Chamber
Huge Chamber
NASA: Glenn Research Center
Moving walls
Large Chamber
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
7
KEY REVERBERATION
CHAMBER
CHARACTERISTICS FOR
WIRELESS MEASUREMENTS
Wireless Applications
•
Radiated power of mobile phones
•
Antenna efficiency measurements
•
Measurements of receiver sensitivity of mobile terminals
•
Investigating biological effects of cell-phone base-station RF exposure
•
Device testing in emulated Rayleigh multipath environments
•
Device testing in emulated Rician multipath environments
•
Gain obtained by using diversity antennas in fading environments
•
Measurements on multiple-input multiple-output (MIMO) systems
8
Standardization of Wireless Measurements
Can we use a reverberation chamber as a reliable and
repeatable test facility that has the capability of simulating
key multipath environments for the testing of wireless
communications devices?
If so, such a test facility will be useful in wireless
measurement standards.
Testing Application: Total Radiated Power
from Cell-Phones
Data from CTIA working group on total radiated 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
Reverb chamber
data has less
variability than
the anechoic data!
3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
Low
Mid
High
Channel
9
Can we model multipath environments in a
reverberation chamber?
Reverberation Chambers are
Natural Multipath Environments
10
Reverberation Chamber Provides a
Statistically Uniform Channel
Power vs. Azimuth Angle
1.5 GHz
2.5 GHz
• At a single frequency, observed received power is not uniformly
distributed over azimuth.
21
Channel is Frequency Dependent
Power vs. Azimuth Angle: Paddle Configuration A
•
•
Strong frequency dependence is indicative of an isotropic scattering environment.
Constructive and destructive interference cause peaks and nulls.
22
11
Power vs. Azimuth Angle: 3.6 MHz BW
1.5 GHz
2.5 GHz
• Received power is distributed more uniformly when observed over
some bandwidth.
23
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:
The Amplitude of E is either Rayleigh or Ricean depending if a LOS path is present.
Urban Environment
Rural Environment
12
Ricean K-factor
K-factor: k 
direct component
scattered components
or K = 10Log(k)
K= -∞ dB
(Rayleigh)
K=1 dB
K=4 dB
K=10 dB
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?
13
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
Calculating the Ricean K-factor in a reverberation chamber [1]
• 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.
•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.
•By varying Q (the chamber quality factor), the K-factor can be changed.
• 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.
14
Measured K-factor for Chamber Loading
1.0E+03
1.0E+02
6 pcs absorber
K-factor
2 pcs absorber
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.
Simulating Propagation Environments
with Different Impulse Responses and RMS Delay Spreads
-13
x 10
-12
10
x 10
7
6
Transmitting
antenna
Receiving
antenna
PDP (linear)
PDP (linear)
8
6
4
62 inches
VNA
Port 1
4
3
2
Rx8
rms = 105 ns
2
tripods
5
0
1000
2000
Delay (ns)
3000
Rx12

= 232 ns
1
0
rms
1000
2000
Delay (ns)
3000
Port 4
Ground Plane
Fiber Optic
Transmitter
RF Optical
200 m
Optical
Fiber
Fiber Optic
Receiver
RF Optical
15
S21 Measurements: Loading the Chamber
Impulse Responses and Power Delay Profiles
Power Delay Profile:
where h(t) is the Fourier transform of S21(w)
Loading the Chamber
16
RMS Delay Spreads from Q measurements
where K is the K-factor and a threshold.
Thus, once we have Q, we can estimate rms
Impulse Responses and RMS Delay Spreads
for Different Ricean K-factors
17
Same Absorber, Different PDPs
Mean Power Delay Profile (100 steps)
Mean Power Delay Profile (100 steps)
-60
-60
Horn Indirect 1, Absorber A
Horn Indirect 1, Absorber A
Horn Indirect 2, Absorber A
-70
Horn Indirect 2, Absorber A
-70
Horn Indirect 1, Absorber B
Horn Indirect 1, Absorber B
Horn Indirect 2, Absorber B
Horn Indirect 2, Absorber B
Horn Indirect 1, Absorber C
Horn Indirect 2, Absorber C
Horn Direct CoPol, Absorber B
-90
Horn Direct XPol, Absorber B
-100
-110
Horns
-120
-130
0
500
Horn Indirect 1, Absorber C
-80
Power Delay Profile (dB)
Power Delay Profile (dB)
-80
Horn Indirect 2, Absorber C
Omni Direct CoPol, Absorber B
-90
Omni Direct XPol, Absorber B
-100
-110
Omnis
-120
1000
Delay (ns)
1500
-130
0
2000
500
1000
Delay (ns)
1500
2000
•Different antenna types can impact PDP, just as absorber can
Instantaneous Results Can Vary
RMS Delay Spread at Each Paddle Position
200
180
160
120
100
80
3 Abs Floor, Horn Pos 1
60
3 Abs Floor, Horn Pos 2
1 Abs Floor, 2 Abs Raised, Horn Pos 1
40
1 Abs Floor, 2 Abs Raised, Horn Pos 2
20
All PDPs, Antenna Position 2a, Abs Ht 2, Xpol
3 Abs Raised, Horn Pos 1
20
3 Abs Raised, Horn Pos 2
0
0
20
40
60
Paddle Position
80
100
18
16
Power Delay Profile (linear units)
RMS Delay Spread (ns)
140
14
12
10
8
6
4
2
0
0
100
200
300
400
500
Delay (ns)
600
700
800
18
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
*Mention of products by name does not imply endorsement
by NIST. Other products may work as well or better.
BER Measurements*
Calibration: determine delay between received external trigger and the first demodulated bit of the
pattern
This offset depends on the signal bit rate.
Measurement: VSG repeatedly transmits an a priori known bit pattern
VSA locks on the external trigger and stores the received bit pattern
VSA integrates the signal spectrum to find the received signal power
LabVIEW: receives the bits from the VSA
corrects for the determined offset
calculates the BER
BER running average - 24.3 ksps BPSK with slow paddle speeds
0.06
VSG transmit power: -65 dBm
VSG transmit power: -55 dBm
VSG transmit power: -45 dBm
0.05
0.04
BER
Each measurement is
finished when the BER
has settled
0.03
0.02
This measurement is repeated
for a range of VSG transmit powers
0.01
0
0
0.5
1
1.5
2
Number of received bits [N]
2.5
3
3.5
6
x 10
*Mention of products by name does not imply endorsement by NIST. Other products may work as well or better.
19
BER Measurements
BER for a 243 ksps BPSK signal
BER for a 786 ksps BPSK signal
CREATING REALISTIC
ENVIRONMENTS FOR
TESTING WIRELESS DEVICES
20
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.
?
Small Chamber = Large Structure
•Omnidirectional antennas
•Instantaneous differences (even if mean characteristics match)
•Monotonic, exponential decay
•Decay time
-12
10
-13
x 10
x 10
7
6
PDP (linear)
PDP (linear)
8
6
4
Rx8
rms = 105 ns
2
0
1000
2000
Delay (ns)
3000
5
4
3
2
Rx12

= 232 ns
1
0
rms
1000
2000
Delay (ns)
3000
21
Example: Denver High-Rise Building Tests
18
19
21
10
15
14
13
12
11
8
5
7
6
3
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
Port 4
Ground Plane
Fiber Optic
Transmitter
RF Optical
200 m
Optical
Fiber
Fiber Optic
Receiver
RF Optical
Can we replicate the Denver high-rise in a
reverberation chamber?
horn
antenna
mode stirrers
wireless
device
antenna
absorber
0
1 absorber: rms=187 ns
-5
-10
3 absorber: rms=106 ns
Power Delay Profile (dB)
-15
-20
7 absorber: rms=66 ns
-25
-30
-35
-40
-45
-50
-55
Reverberation chamber with absorbing material
and phantom head
-60
Large office biulding
rms=59 ns
-65
-70
0
50
100
150
200
250
300
time (ns)
350
400
450
500
22
Can the reverberation chamber simulate an oil
refinery?
Power Delay Profile in an oil refinery.
Lab-Based Test Method
Location and
Notes
Test Point
VNA Loss
Data (dB)
Path Loss
@700 MHz
(dB)
RMS Delay
Spread @700
MHz
(ns)
Republic
Notes:
- System 1
repeater at test
point 2.
1
7.23
68.6149
44.99
2
27.06
88.4449
39.52
3
38.15
99.5349
52.30
4
37.60
98.9849
133.41
5
37.18
98.5649
81.25
6
42.26
103.6449
102.78
7
46.04
107.4249
138.29
8
44.88
106.2649
104.69
9
48.30
109.6849
376.10
10
45.34
106.7249
338.17
11
50.25
111.6349
167.91
12
50.48
111.8649
231.57
13
50.98
112.3649
209.07
14
51.82
113.2049
192.25
15
49.60
110.9849
240.20
16
44.64
106.0249
377.45
17
29.28
90.6649
296.87
18
30.45
91.8349
161.75
19
42.24
103.6249
429.90
20
39.30
100.6849
333.25
21
47.07
108.4549
453.47
18
19
21
10
15
14
13
12
11
8
5
7
6
3
20
17
16
9
4
North
1
West
2
East
South
VNA measurement test locations are in pink
Compare wireless device performance measured in the field to performance in the
reverberation chamber
23
Adding More Realism to the PDP
-90
-100
-110
Noise Threshold = -116 dB
-120
PDP (dB)
-120
-130
-140
-150
-140
-150
TX1 to RX9
rms = 39 ns
200
400
600
800
Delay (ns)
1000
1200
R9
Noise Threshold = -116 dB
-170
0
500
Delay (ns)
1000
R10
R5
•Mean of 27 NLOS
measurements made in Denver
urban canyon.
T1
T
R1212
T3
T
T2
•Channel characterization and
PASS device measurements.
Clustering of Multipath is Common
Shortest path
1
fitted simulated data
measured Data Denver
0.9
TX
0.8
0.7
0.6
0.5
0.4
0.3
RX
0.2
Welton Street
0.1
0
RX
1
0
100
200
300
400
500
600
Delay [ns]
Mean of 27 NLOS
measurements
700
800
900
RX
2
1000
Parking lot
17 th Street
-170
0
-130
-160
-160
PDP
PDP (dB)
TX1 to RX5

= 115 ns
rms
-110
RX
3
Clusters of
exponential
distributions off of
buildings
TX
Transmitter Site
Glenarm Pl
24
Channel Emulator (CE) and Vector Network
Analyzer (VNA) used with a Reverberation
Chamber
Generate a Complex Power Delay Profile [8]
25
REVERBERATION CHAMBER
TEST ENVIRONMENTS 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
TX
RX
MISO
TX
RX
SIMO
TX
RX
MIMO
TX
RX
26
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
GPIB
RF out
LO in
LO out
I in
Q in
Event 1
IEEE 1394
Firewire
VSG 2
Pattern sync.
VSA
RF out
•Error correction only to recover
constellation after deep fade
RF in
LO in
•Paddle stepped
Reverberation chamber
Stirrer
INIT
Initialize the stirrer
Initialize the VSA
Test Methodology
Beamforming?
•One receiver is used: repeatability and
post processing are key
•Careful synchronization of
generators, receiver, paddles required
•Post processing implements diversity,
MIMO schemes
•System is automated to prompt user
for correct inputs
•System applicable for design (laptop
antennas, for example) and test of
algorithms
Yes
No
Beamforming
Shifting the phase until
maximum power is reached
at the receiver
Recording data
Recording data
Stepping the stirrers
Stepping the stirrers
All steps done?
No
No
Yes
Yes
Switch antenna!
Switch antenna!
Tells the user to switch to a
different antenna
Tells the user to switch to a
different antenna
All antennas
used?
All steps done?
No
No
Yes
All antennas
used?
Yes
Storing the data to file
Exiting the program
27
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
SISO
Real life
Simulation in reverberation chamber
Stirrer
Reverberation
chamber
Transmitter
Receiver
Vector
Signal
Generator
•No choice of received signal
Vector
Signal
Analyzer
•Antenna orientation sets level of
direct vs. reflected signal in the
reverb chamber
28
SIMO
Real life
Simulation in reverberation chamber
A
Stirrer
B
A
Reverberation
chamber
Transmitter
B
Receiver
Antenna A receives strongest
signal
•Also called Diversity
•Power meter monitors
received signal strength
•Strongest signal
demodulated by receiver
Vector
Signal
Generator
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
TX BEAMFORMING
Real life
Simulation in reverberation chamber
Stirrer
Phase
Reverberation
chamber
Transmitter
Receiver
Signal strength
Vector Signal
Generator 1
Phase
•TX antennas phase adjusted
to maximize received signal
•Received signal strength
relayed to transmitter
•Strongest signal demodulated
by receiver
Vector
Signal
Analyzer
Signal strength
Vector Signal
Generator 2
•VSGs are phase locked
• Phase swept 360°
•Strongest channel power
chosen in post processing
29
RX BEAMFORMING
Real life
Phase
Not implemented in the
simulation yet
Transmitter
Receiver
•Phasing of RX antennas
adjusted based on received
signal strength
•Strongest signal demodulated
by receiver
MIMO
Real life
Simulation in reverberation chamber
Stirrer
A
B
Transmitter
C
Reverberation
chamber
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
30
MIMO
Real life
Simulation of various MIMO
schemes:
Simulation in reverberation chamber
Stirrer
A
B
C
Reverberation
chamber
D
Receiver
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
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
Frequency = 2.4 GHz
Pout = -50 dBm
BPSK modulated signal
MIMO without
beamforming
0
768 1 Msps 2.5 3 Msps5 Msps 10
ksps
Msps
Msps
25
20
BER [%]
Transmitter
•Precoding: TX beamforming
used.
•Spatial multiplexing: TX data
start time not sufficiently precise
for simultaneous transmit. One
TX, one RX used in multiple
combinations. Data recombined in
post processing.
•Diversity coding: One RX so
only one code implemented. VSA
reports channel power, strongest
path chosen.
TX beamforming (SIMO)
15
Receive diversity with TX
beamforming
10
MIMO
5
0
768
1
2.5
3
5
10
ksps Msps Msps Msps Msps Msps
31
Number of Absorbers
5
4.5
BER SISO
4
BER MISO
BER [%]
3.5
BER SIMO
3
BER MIMO no Beamforming
2.5
BER MIMO
2
1.5
1
0.5
12
11
10
9
8
7
6
5
4
3
2
1
0
Number of absorbers inside chamber
Frequency = 2.4 GHz
Pout = -50 dBm
BPSK modulated signal, 768 ksps
70%
Non-line-of-sight: Nested Reverberation Chamber
Line of sight,
no nested chamber
Line of sight, with
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
32
Nested Chamber Wireless Environment
0
K-Factor, (dB)
-3
-6
-9
-12
LOS, w/o nested chamber
LOS w/ nested chamber, small paddle
NLOS w/ nested chamber, w/o small paddle
NLOS w/ nested chamber, w/ small paddle
-15
-18
-21
1
1.5
2
2.5
300
4.5
5
5.5
6
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
250
RMS Delay Spread (ns)
3
3.5
4
Frequency, (GHz)
200
150
100
50
mean K values: -4.3
-5.3
0
20
-6.9 -10.7
0
10
30
40
Threshold (dB)
50
60
70
80
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
Network
Analyzer
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
| 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
33
Correlation in the Nested Chamber MIMO Channels
(2.4 GHz)
With small paddle stirring
Without small paddle stirring
S13
S13
S14
S14
S23
S23
S24
S24
S31
S31
S32
S32
S41
S41
S42
1
0.9
0.8
0.7
S42
0.6
S13 S14 S23 S24 S31 S32 S41 S42
S13 S14 S23 S24 S31 S32 S41 S42
With small paddle stirring
and one absorber
With small paddle stirring
and two absorbers
0.5
S13
S13
S14
S14
S23
S23
S24
S24
S31
S31
S32
S32
S41
S41
S42
S42
S13 S14 S23 S24 S31 S32 S41 S42
0.4
0.3
0.2
0.1
S13 S14 S23 S24 S31 S32 S41 S42
0
Reverberation Chamber Standards Proposed
Testing Methods
Standards
•International Standard IEC 61000-4-21: Testing and
measurement techniques – Reverberation chamber test methods
•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]
34
Challenges and Opportunities in Reverberation
Chamber Wireless Testing
•
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
Conclusions and Observations
1. Reverberation chambers represent reliable and repeatable test facilities that have
the capability of simulating any multipath environment for the
testing of wireless communications devices.
2. Such a test facility will be useful in the testing of the operation and
functionality of the new emerging wireless devices in the future.
3.
Such a test facility will be useful in wireless measurements standards.
•
•
•
•
•
•
•
•
•
Radiated power of mobile phones
Receiver Sensitivity
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
35
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.
36