Ts7124otaapplicationnote1619630454350
Ts7124otaapplicationnote1619630454350
Ts7124otaapplicationnote1619630454350
TS7124M
5G NR FR1 OTA
Application Note
Products:
ı R&S®TS7124M ı R&S®CMWC
ı R&S®CMX500 ı R&S®CMWflexx
ı R&S®CMW500 ı R&S®CMsquares
This application note describes the procedure to achieve maximum data throughput measurement over-
the-air (OTA) using an RF Shielded Box TS7124M, with a focus given to 5G NR FR1.
Claude LE R OUX
1.2020 – 01.00
Application Note
COMP ANY RESTRICTED Table of Contents
Table of Contents
1 Introduction ......................................................................................... 4
2 Theory .................................................................................................. 5
2.1 NR Maximum Data Rate ..............................................................................................5
2.1.1 Aggregated component carriers ....................................................................................5
2.1.2 MIMO layers ...................................................................................................................5
2.1.3 Modulation Order and MCS ...........................................................................................5
2.1.4 Scaling Factor ................................................................................................................6
2.1.5 Rmax ..............................................................................................................................6
2.1.6 Physical Resource Block and Bandwidth ......................................................................6
2.1.7 Numerology....................................................................................................................6
2.1.8 Symbol duration (Ts)......................................................................................................6
2.1.9 Overhead (OH) ..............................................................................................................6
2.2 Other Parameters .........................................................................................................7
2.2.1 Time Slots ......................................................................................................................7
2.2.2 DMRS ............................................................................................................................7
2.2.3 BLER ..............................................................................................................................7
2.2.4 CSI .................................................................................................................................7
2.2.5 CQI .................................................................................................................................8
2.3 Expected Maximum NR Throughput ..........................................................................8
10 Conclusion ........................................................................................ 36
1 Introduction
Maximum data throughput has always been an important Key Performance Indicator
(KPI) for most wireless companies.
That has been further emphasized by 5G NR, which allows even greater data rate.
In the past, testing was mainly done in conducted mode, thanks to tiny U.FL RF
connectors available on commercial phones.
However, most mobile phone manufacturers recently got rid of those RF connectors
from their commercial devices, therefore preventing testers to proceed with conducted
mode unless soldering RF connectors to Device Under Test (DUT).
As a result, the importance of OTA chambers and shielded boxes, such as the
TS7124M, will further increase.
Besides, the TS7124M offers flexibility in terms of RF antennas (type, number and
location) and scenarios (multi-cells, carrier aggregation, MIMO Schemes).
This application note describes the procedure to achieve maximum data throughput
measurement over-the-air (OTA) using an RF Shielded Box TS7124M, with a focus
given to 5G NR FR1.
2 Theory
2.1 NR Maximum Data Rate
3GPP TS 38.306-4.1.2 shows how the maximum data rate is computed:
BW ( j ),
12
J
v
N PRB
data rate (in Mbps) 10 6 ( j)
Qm( j ) f ( j)
Rmax 1 OH ( j )
j 1
Layers
Ts
Useful tool: https://5g-tools.com/5g-nr-throughput-calculator/
Throughput depends then on:
▪ number (j) of aggregated component carriers (CC)
▪ number (v) of MIMO layers
▪ modulation order (Qm)
▪ scaling factor (f)
▪ Rmax
▪ Resource Block (RB) allocation (N)
▪ numerology (µ)
▪ symbol duration (Ts)
▪ overhead (OH)
In order to get maximum throughput, one should target maximum modulation order
(Qm) and maximum modulation coding scheme (MCS).
Qm:
ı 2: QPSK
ı 4: 16 QAM
ı 6: 64 QAM
ı 8: 256 QAM
As per 3GPP TS 38.214 (v.15.8.0, table 5.1.3.1), for downlink:
ı MCS table 1 should be considered if MCS-Table is set to 'qam64'
2.1.5 Rmax
2.1.7 Numerology
10 3
Ts
14 2
Here: Ts = 35.7 µs (including cyclic prefix)
2.2.2 DMRS
2.2.3 BLER
A very good physical radio link is critical to get throughput at any protocol layer.
Block Error Rate (BLER) is a good criterion to evaluate the radio link quality.
In order to get a maximum measured throughput value very close to the scheduled and
expected throughput value, the target BLER threshold is set here to 0.01.
I.e. BLER must be lower than 1% so that measured throughput value is retained as
candidate for supported maximum data rate.
2.2.4 CSI
2.2.5 CQI
As part of CSI report, Channel Quality Indicator (CQI) gives the expected MCS based
on downlink channel estimation made by the UE, i.e. best MCS without exceeding
given target BLER.
3GPP TS 38.214 defines three CQI tables.
3 Needed Equipment
3.1 Hardware
Hardware equipment required for performing OTA testing for NSA FR1 are:
► 1 x DUT holder R&S®TS-24P1 with 2 positions on the door (only 1 possible if full
antenna ring is used though)
► 2 x Z24 RF combiners for multi-cell MIMO 4x4 cases (such as NSA MIMO 4x4
here)
► 1 x Smartphone referred to as DUT. More than one DUT was used for testing, but
the DUT best results were achieved with is referred to as "Reference DUT"
3.2 Software
ı CMsquares
ı XLAPI
▪ PyCharm, Visual Studio or any other python IDE
4 TS7124M Setup
This section provides you details on how to configure and set up the TS7124M. There
can be more than one way of placing the Vivaldi antennas and DUT in a shielded box.
The optimal position depends on the antenna location within the DUT, which is not
always known.
In this section we explain the configuration that has provided most optimal and
maximum throughput. Hereafter, we refer to this as SETUP1.
There were other setups that were also tested and evaluated. Details of these can be
found in Appendix E and Appendix F.
4.1 RF cabling
The TS7124M comes by default with 4 x N (outside) to SMA (inside) RF feedthrough
connectors at the rear of the shielded box.
For LTE SISO + NR MIMO 4x4 scenario, a single low insertion loss (IL=0.7dB) 2-1 port
RF combiner has been used to combine LTE and NR main paths to the same Vivaldi
antenna, as a first evaluation.
Then, a single Z24 RF combiner has been evaluated for the same scenario.
Finally, for NSA MIMO 4x4 scenario, a pair of Z24 RF combiners has been used
(referred here as “TOP” and “BOT(tom)”) to combine NR and LTE paths as detailed
below:
As you may see, rails are oriented to get a multiple of 45° as angle of rotation between
Vivaldi antennas for polarization diversity.
Below shows Vivaldi antenna position, orientation and internal TS7124M SMA RF
connector antennas are connected to:
The configuration allows to simply "surround" the DUT with Vivaldi Antennas.
Other antenna configurations are considered in Appendix E.
Source: https://www.rfglobalnet.com
Therefore, the DUT screen will face down (back of the smartphone facing the Vivaldi
Antennas).
Four positions have been considered enough to evaluate antenna location effect on
throughput measurement.
Position 1: DUT top directed towards the back wall.
Position 2: DUT top directed towards the right wall.
Position 3: DUT top directed towards the front wall.
Position 4: DUT top directed towards the left wall.
For each run, the DUT will then be rotated manually by 90° clockwise on the tray
(horizontal plane).
In order to "print" the possible DUT positions, it is recommended to cut 2 pieces of
paper of the DUT size and tape them at the back of the DUT holder to finally get a
"cross" shape:
Position 1
Position 2
Position 4
Position 3
You may then tape DUT and USB cable onto the DUT holder to prevent the DUT to
move when opening / closing the door.
5 System Settings
5.1 External RF Attenuation
For RF systems, calibration is usually mandatory. However, here, for throughput
evaluation, it should not be required. Specific calibration procedure, reference antenna
and probes are not needed.
For now, an external RF attenuation of 25 dB has been arbitrary considered for all
setups as reference for simplicity, whatever the RAT and the band (refer to Appendix C
for details).
Settings in R&S CMsquares (e.g. for NSA MIMO 4x4 N78 B1):
1. Test Environment -> Network -> Predefined Network -> LTE 4x4, NR 4x4
2. LTE Network Configuration
Parameter Value
Duplex Mode FDD (default)
Frequency Band Indicator 1 (default)
DL & UL: Resource Blocks 100 / Frequency Bandwidth: 20
MHz
Range Choice Mid Band (default)
Max. Cell Power - 57.2 dBm (default)
MIMO Scheme MIMO 4x4
3. NR Network Configuration
Parameter Value
Frequency Range FR 1 (default)
Subcarrier Spacing 30 kHz (default)
Duplex Mode TDD (default)
Frequency Band Indicator N 78 (default)
Carrier Bandwidth 100 MHz (default)
Total Cell Power -44.8 dBm (default)
ss-PBCH-BlockPower 20 dBm
# DL Slots 8
# UL Slots 1
MCS Table 256 QAM
Slot Assignment Configuration (0 - 7)
# RB 273
MIMO Scheme MIMO 4x4
No DMRS optimization add_pos = 2 (default)
Parameter Value
Transmission Mode TM3 or TM4
DCI Format DCI 2
Start RB 0
Number RB 100
MCS 26
NOTE: A SCPI script may be done to automatically increase MCS in order to quickly
evaluate maximum throughput, and be launched via FORUM (Appendix G). So far,
such task has been done via XLAPI test script.
6.1 Summary
Results show maximum throughput is achieved with reference DUT for SETUPs 1, 3
and 4 (details in Appendix F).
However, testing with another DUT reveals best throughput is reached with SETUP 1:
When using SETUP 1, results show maximum throughput is achieved with reference
DUT in any position:
Position 2 27 0 1.53
Position 3 27 0 1.53
Position 4 27 0 1.53
However, testing with another DUT reveals best throughput is reached when DUT is in
position 1:
8.1 LTE
MIMO 4x4, 100 RBs / 20 MHz, TM 3/4, DC2 format, 256QAM table (Appendix H)
Total Power = -50 dBm
8.2 NR
Default Total NR Cell Power (-44.8 dBm):
9 NR Throughput at IP level
Once maximum MAC throughput and good BLER are reached, IP throughput may be
evaluated.
iPerf, a well-known open-source network testing tool, is used to do so.
Two main protocols provide end-to-end communication between network entities
(client and server):
ı Transmission Control Protocol (TCP)
ı User Datagram Protocol (UDP)
TCP is:
ı Connection-oriented; a connection has to be set before transmitting TCP data
(stream of bytes) from a single transmitter to a single receiver,
ı Reliable; providing congestion, order and error-checking mechanisms.
UDP is:
ı Connectionless; UDP data (datagrams) can be transmitted as either IP broadcast
or multicast,
ı Unreliable; prioritizing time over reliability,
Therefore, TCP is well suited for many internet applications such as web browsing,
emails and file sharing, while UDP is suitable for real-time applications such as Voice
over IP (VoIP).
NOTE: Links to all required third-party tools are available in User Manuals and Tools
section for download.
Below shows iPerf results on the Server side (DUT) for DL TCP, via an iPerf
application (e.g. Magic iPerf) and a mirror application (e.g. Scrcpy, Vysor):
Settings:
ı RLC AM (Acknowledge Mode)
ı iPerf settings:
▪ on Tester (CMsquares/XLAPI); equivalent to iperf -c 172.22.1.100 –w
10240K –i 1 –p 5010 –P 2:
▪ Application Type: iPerf2
▪ Direction: client (-c)
▪ Server IP address: 172.22.1.100 (DUT)
▪ TCP window size (-w) = 10240 kB (10 MB, but value is expected in kB
by WebGUI). NOTE: Specifying a large window size reduces the amount
of processing that needs to be done on both client and server sides.
▪ Interval (-i) = 1 s
▪ Port (-p) = 5010. NOTE: default port “5001” is sometimes not available
so other port id should be considered instead to avoid any eventual
conflict.
▪ Number of parallel threads (-P) = 2. NOTE: high data throughput may
require parallel streams to get maximum channel load (1.2 Gbps max
reached here with a single thread otherwise).
Optional:
▪ Buffer length (-l) = 128 kB by default for TCP – packet size.
When testing, if UDP throughput is found limited per thread, more than one iPerf
instance will have to be launched from both client and server and with different port for
each thread.
On DUT side, iPerf android applications (e.g. Magic iPerf) do not provide the capability
to run more than one iPerf instances. NOTE: only one instance of the same android
application can be launched.
Therefore, iPerf instances will have to be run from different shell / command prompts
(on PC via adb – to avoid USB tethering – or directly from the DUT via e.g. “Terminal
Emulator for Android” application).
Procedure to run iPerf commands via adb:
1. Install adb on PC
2. Set the DUT to Developer mode (multiple taps on build number in settings),
enable USB debugging, “always allow USB debugging” when popped up and
reboot it.
3. From PC (adb repository):
adb devices (to make sure device is seen)
adb push iperf_2_0_13a /data/local/tmp/iperf
adb shell chmod 755 /data/local/tmp/iperf
adb shell /data/local/tmp/iperf –s –u –i 1 –p 5011 (/5012/5013/5014
from different prompts)
NOTE: Launching adb commands might conflict with Scrpy / Vysor so you might have
to relaunch the application later if required.
Settings:
ı RLC AM (Acknowledge Mode); similar results found with RLC UM
(Unacknowledged Mode) though.
ı iPerf settings
▪ On Tester (CMsquares/XLAPI), iPerf settings (equivalent to iperf -c
172.22.1.100 –u –b 350M –i 1 –p 5011 (/5012/5013/5014)):
▪ Application / Type: iPerf2
CMsquares:
DUT:
iPerf instance 1:
iPerf instance 2:
iPerf instance 3:
iPerf instance 4:
10 Conclusion
ı Testing results reveal NR maximum data throughput can be reached OTA with
TS7124M.
ı Results also show throughput KPI can be evaluated even when DUT is in near
field region.
ı TS7124M is well suited for Signaling Procedure verification and Throughput
testing.
ı As a reminder, please find here below the best setup found for smartphone testing
with:
▪ Vivaldi antenna type, position and orientation
▪ DUT position and orientation
20
13
13
ı You may also test with other DUT types like laptop, there is enough room in
TS7124M for that.
ı If best performance for laptop is not achieved with the above setup, please refer to
other setups described in Appendix E.
ı Few troubleshooting tips:
▪ make sure significant external RF attenuation is set
▪ if BLER is higher than expected, increase DL power
▪ change one parameter at a time for first evaluation
ı If interested in controlling remotely the opening of the shielded box, you might
wish to consider the TS7124AS variant instead, more details in Appendix I.
12 Ordering Information
Here below is an example of a basic TS7124M configuration you may start with:
To get access to all provided options, you may refer to TS7124M product brochure:
https://scdn.rohde-
schwarz.com/ur/pws/dl_downloads/dl_common_library/dl_brochures_and_datasheets/
pdf_1/TS7124M_bro_3607-3839-12_v0200.pdf
Parameter Value
Parameter Value
Impedance 50 Ω
Source: https://www.comsol.com/blogs/vivaldi-antenna-design-analysis/
C Appendix C: Free Space Pathloss Equation
Near Field region: close to the antenna. EM fields are somewhat unpredictable so no
measurements are usually made in this region, further divided in two parts:
ı Reactive Near Field: electric E-field and magnetic H-field are 90 degrees out of
phase with each other (reactive). Reminder: to radiate (propagate), E/H fields
need to be orthogonal and in phase with each other.
ı Radiative Near Field (also known as "Fresnel Region"): EM fields start to
transition from reactive to radiating fields. However, the shape of the radiation
pattern still varies with distance.
Far Field Region (also known as "Fraunhofer region"): far from the antenna. E and H-
fields are orthogonal to each other and to the direction of propagation as with plane
waves.
Antennas are usually used to transfer signals at large distances which are considered
to be in the far-field region. One condition that must be met when making
measurements in the far field region is that the distance from the antenna must be
much greater than the size of the antenna and the wavelength.
D: maximum linear dimension of the antenna.
λ: wavelength of EM wave
Here:
Vivaldi antenna size (D) = 12 cm
SETUP 2
This setup configuration allows the DUT to get the farthest position from Vivaldi
Antennas to attempt to be in Far Field area (Appendix D for details).
Same considerations as SETUP 1 must be taken into account excepting:
▪ the DUT holder tray is removed from the front door
▪ a half antenna ring is recommended
In order to place the DUT directly at the bottom of the TS7124.
SETUP 3
This setup configuration improves polarization diversity by placing the Vivaldi Antennas
located on the top middle rail 90° apart.
The DUT is placed at the bottom of the TS7124.
SETUP 3a
Same antenna positions as SETUP 1, i.e.:
▪ 6 & 20 on top middle rail
▪ 13 on side rails
SETUP 3b
"Widest" possible antenna positioning in SETUP 3 configuration, with following
antenna positions:
▪ 6 & 24 on top middle rail
▪ 15 on side rails
SETUP 3c
This setup is a variant of SETUP 3 with a R&S®TS-F24-V3 Cross-Polarized Vivaldi
Antenna V3 (1.7 - 20 GHz) on the top middle rail (position 13).
Its frequency range does not go as low as R&S®TS-F24-V1 (0.7 - 14 GHz) so could be
an issue, depending on the band used.
SETUP 4
This setup configuration improves polarization diversity by placing pairs of Vivaldi
Antennas 90° apart (same as for SETUP 3a) and the DUT on the tray (same as for
SETUP 1).
SETUP 5
This setup configuration requires a full antenna ring.
It further improves polarization diversity by placing all 4 Vivaldi Antennas 90° apart (2
on top rails, 2 on bottom rails) and placing the DUT on the tray (as for SETUP 1).
However, 2 Vivaldi antennas, placed at the bottom, will then face the DUT screen…
Vivaldi antenna position: 13.
SETUP 6
This setup configuration uses 2 x R&S®TS-F24-V3 Cross-Polarized Vivaldi Antenna
V3 (1.7 - 20 GHz) placed on the top side rails and the DUT on the tray.
F Appendix F: Results of other tested TS7124M SETUPs
By getting both Vivaldi antennas as close as possible (positions 8 & 18), performance
degrades:
By getting both Vivaldi antennas further apart (SETUP 3b), performance degrades:
Right now, WebGUI does not provide the capability to set any external RF attenuation
value, and so has to be set either via XLAPI or SCPI script.
FORUM is a free-of-charge SW provided by R&S that allows to send SCPI commands.
1. Download FORUM from:
https://scdn.rohde-
schwarz.com/ur/pws/dl_downloads/dl_application/application_notes/1ma196/RS_
Forum_Setup_Win_3_3_6.exe
Details from: https://www.rohde-schwarz.com/au/applications/using-r-s-forum-
application-for-instrument-remote-control-application-note_56280-50946.html
pathLossTableMainExist =
cmw.ask("CONF:BASE:FDC:CTABle:EXISt? 'pathLossTableMain'")
if pathLossTableMainExist:
cmw.write("CONFigure:BASE:FDCorrection:CTABle:DELete
'pathLossTableMain'")
pathLossTableDivExist = cmw.ask("CONF:BASE:FDC:CTABle:EXISt?
'pathLossTableDiv'")
if pathLossTableDivExist:
cmw.write("CONFigure:BASE:FDCorrection:CTABle:DELete
'pathLossTableDiv'")
cmw.write("CONFigure:BASE:FDCorrection:CTABle:CREate
'pathLossTableMain', 700000000 , 25, 5500000000 , 25")
cmw.write("CONFigure:BASE:FDCorrection:CTABle:CREate
'pathLossTableDiv', 700000000 , 25, 5500000000 , 25")
TX_connectorsMain = ['R14C','R24C']
RX_connectorsMain = ['R14C','R24C']
TX_connectorsDiv =
['R11C','R13C','R12C','R21C','R23C','R22C']
RX_connectorsDiv =
['R11C','R13C','R12C','R21C','R23C','R22C']
CAUTION: each reset via CMsquares will remove those attenuation settings.
H Appendix H: LTE 256QAM MCS Table
I Appendix I : TS7124AS
R&S also provides a variant of the TS7124M: the TS7124AS, which provides an
automatic drawer opening thanks to a compressor connected to it.
It is intended more for production rather than R&D, hence why we referred to TS7124M
through the current document.
Besides a different opening (automatic vs manual), the ceiling is screwed in the
TS7124AS (rather than clipped on the TS7124M) as per the picture below.
B
BLER: Block Error Rate
C
CC: Component Carrier
CMW500: Wideband Radio Communication Tester
CMX500: 5G NR Radio Communication Tester
CMWC: CMW500 Controller
CQI: Channel Quality Indicator
CSI: Channel State Information
D
DL: Downlink
DMRS: Demodulation Reference Signal
DUT: Device Under Test
F
FF: Far Field
I
IPERF: Internet Performance Working Group
K
KPI: Key Performance Indicator
M
MIMO: Multiple Input Multiple Output
MCS: Modulation Coding Scheme
MSS: Maximum Segment Size
MTU: Maximum Transmission Unit
N
NA: Not Applicable
NF: Near Field
NR: (5G) New Radio
O
OH: Overhead
OTA: Over The Air / over-the-air
P
PC: Personal Computer
PDU: Protocol Data Unit
Q
Qm: Modulation Order
R
RS: Reference Signal
RSRP: Reference Signal Received Power
RB: Resource Block
RF: Radio Frequency
S
SSB: Synchronization Signal Block
T
TCP: Transmission Control Protocol
TM: Transmission Mode
TS: Time Slot
U
U-FL: Series-name of a particular RF connector, made by Hirose
UDP: User Datagram Protocol
UL: Uplink
Rohde & Schwarz Regional contact
The Rohde & Schwarz electronics group offers Europe, Africa, Middle East
innovative solutions in the following business fields: +49 89 4129 12345
test and measurement, broadcast and media, secure customersupport@rohde-schwarz.com
communications, cybersecurity, radiomonitoring and
North America
radiolocation. Founded more than 80 years ago, this
1 888 TEST RSA (1 888 837 87 72)
independent company has an extensive sales and customer.support@rsa.rohde-schwarz.com
service network and is present in more than 70
countries. Latin America
+1 410 910 79 88
The electronics group is among the world market
customersupport.la@rohde-schwarz.com
leaders in its established business fields. The
company is headquartered in Munich, Germany. It Asia Pacific
also has regional headquarters in Singapore and +65 65 13 04 88
Columbia, Maryland, USA, to manage its operations customersupport.asia@rohde-schwarz.com
in these regions.
China
+86 800 810 82 28 |+86 400 650 58 96
customersupport.china@rohde-schwarz.com
Schwarz website.
Version 01.00 | R&S®5G NR FR1 OTA
www.rohde-schwarz.com