Real World Receiver Testing and the 1dB Criteria
17
Real World Receiver Testing and the 1dB Criteria Guy Buesnel, Mark Hunter, Charlotte Perry December 2018
Transcript of Real World Receiver Testing and the 1dB Criteria
Real World Receiver Testing
and the 1dB Criteria
Guy Buesnel, Mark Hunter, Charlotte Perry December 2018
2
Introduction Measurement of GNSS Receiver parameters
• Last year we examined measurement parameters when GNSS
Receivers were subjected to Adjacent band interference
• 1dB degradation in CNR used in European ETSI standard for ABC
– Established Interference Protection Criterion (USAF justification paper)
– Some agencies have proposed use of GNSS accuracy or Time to First Fix as a
preferable metric
• The 1dB IPC does not require the level of harmful RF interference to be
specified..
• Spirent approach – use automated test bench to carry out many
repeated measurements on sample receivers across increasing
interference levels
3
Test Set-up PNT Test Bench
RF Simulator
Scenario Generation
RF Navigation Data
Control
commands
Control
commands
Receiver
Data
GNSS
Device under test
DUT Receivers A,B,C,D,E,X
4
Test Method The GNSS RED Adjacent Band Interference test
Frequency band (MHz) Test point centre
frequency (MHz)
Adjacent frequency
signal power level (dBm)
Comments
1518 - 1 525 1 524 -65 MSS (space-to-Earth) band
1 525 - 1 549 1 548 -95 MSS (space-to-Earth) band
1 549 - 1 559 1 554 -105 MSS (space-to-Earth) band
1 559 - 1 610 GUE RNSS band under test
1 610 - 1 626 1 615 -105 MSS (Earth-to-space) band
1 626 - 1 640 1 627 -85 MSS (Earth-to-space) band
Frequency band (MHz) Test point centre
frequency (MHz)
Adjacent frequency signal
power level (dBm)
Comments
960 - 1 164 1 154 -75 AM(R)S, ARNS band
1 164 - 1 215 GUE RNSS band under test
1 215 - 1 260 GUE RNSS band under test
1 260 - 1 300 GUE RNSS band under test
1 300 - 1 350 1 310 -85 Radiolocation, ARNS, RNSS
(Earth-to-space) band
Parameter Value Comments
Frequency See tables 4-2 and 4-3
Power level See tables 4-2 and 4-3
Bandwidth 1 MHz See clause B.1 for details
Format AWGN
RED: “Radio Equipment Directive”
GNSS RED: ETSI EN 303 413
Detail of requirements and tests necessary for
GNSS receivers
Note: The GNSS RED also includes an emissions
test – not covered in this presentation…….
5
RFI effects on CNR and Position Manual Analysis 5 datasets, 60 samples
-2.5
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
2.5
-120 -100 -80 -60 -40 -20 0
0
5
10
15
20
25
30
35
40
45
50
horizonta
l positio
n e
rror
(m)
RFI power (dBm)
CN
R (
dB
-Hz)
RFI Power and Horizontal position error at 1554MHz with measurement range (HPE) added - Receiver A
(dB-Hz) Pos Error (m)
RED test level CNR drop >1dB
• Monitoring CNR
and Horizontal
Position Error with
increasing RFI
• Full Range error
bars displayed
• RED Test level and
point of 1dB CNR
drop marked out
6
RFI effects on CNR and Position
• CNR and Horizontal
Position error plotted
for all Receivers with
increasing RFI
• Note that CNR drop
off (for all 5 TURs)
occurs before HPE
starts to fluctuate..
• Some of the TURs
clearly much more
susceptible to the
RFI than others….
Manual Analysis 5 datasets, 60 samples, All receivers
-120 -100 -80 -60 -40 -20 0
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
0
5
10
15
20
25
30
35
40
45
50
-120 -100 -80 -60 -40 -20 0
Axis Title
HO
RIZ
ON
TA
L P
OS
ITIO
N E
RR
OR
(m
)
CN
R (
dB
-Hz)
RFI POWER
Horizontal Position Error and CNR changes with increasing RFI
A CNR B CNR C CNR D CNR E CNR A POS B POS C POS D POS E POS
7
RFI effects on CNR and Position
• Note Receiver A HPE
fluctuations once CNR has
degraded to minimum value of
around 22.5 dB-Hz
In-Band interference at GPS L1 Frequency
-5
0
5
10
15
20
-120 -100 -80 -60 -40 -20 0
0
5
10
15
20
25
30
35
40
45
50
HP
E (
m)
RFI Power (dBm)
CN
R (
dB
-Hz)
RFI Power and Horizontal position error at 1575.42MHz, In Band - Receiver A
(dB-Hz) Pos Error (m)
8
RFI effects on CNR and Position
• Interesting comparison
of receiver behaviour –
for both out of band and
in-band interference
CNR drop-off seen to
precede degradation in
HPE values…
In-band and out of band interference comparison
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.0
-120 -110 -100 -90 -80 -70 -60 -50 -40 -30 -20
0
5
10
15
20
25
30
35
40
45
50
HP
E (
m)
RFI Power (dBm)
CN
R (
dB
-Hz)
1533MHz and GPS L1 RFI – Receiver A
InBand SNR 1554 SNR InBand POS 1554 POS
9
HPE and TTFF under RFI TTFF and HPE Comparison – Test Receiver A
-10.00
-5.00
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00
500.00
1000.00
1500.00
2000.00
2500.00
3000.00
-70 -60 -50 -40 -30 -20 -10 0
TTFF
(s)
RFI Power (dBm)
TTFF and HPE under LTE RFI (1533 MHz)
TTFF Initial Error Av Error
HP
E (m
)
• We see that (for this
receiver –and all
others we’ve tested
so far) the HPE
increases before any
degradation in TTFF
10
HPE and TTFF under RFI TTFF and HPE Comparison – Test Receiver A
-10.00
-5.00
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0.00
200.00
400.00
600.00
800.00
1000.00
1200.00
1400.00
1600.00
-65.0 -60.0 -55.0 -50.0 -45.0 -40.0 -35.0 -30.0 -25.0 -20.0
Po
siti
on
Err
or
(m)
TTFF
(s)
RFI
TTFF and HPE under 1554MHz RFI - Receiver Y
TTFF Initial Error Av Error
• Whilst CNR not plotted
for these two tests, we
saw that CNR was a
precursor to increasing
HPE on this receiver
11
Variability in GNSS measurements Example – 3D positioning accuracy
• Receiver “X” – this time 800
measurements taken using
automated test bench
• Probability distribution can now
be modelled and monitored
across interference levels…
• Too few measurements could
be problematic….
12
Variability in GNSS measurements Example – Time To First Fix
-144 dBm
-30dBm
Receiver “X” with 1533MHz
interferer (simulating LTE base
station)
30 Measurements taken using
automated Test Bench
• Any evidence of anomalous
behaviour?
13
Variability in GNSS measurements Example – Time To First Fix data
-144 dBm
-30dBm
Receiver “X” with 1533MHz
interferer (simulating LTE base
station) – Tester makes 3
measurements at each power
level…
3 measurements
3 measurements
“Unlucky” – this small set of
measurements might result in incorrect
conclusions/decisions being made…..
14
Spirent Insights …and further work
• Our results serve to confirm that under test conditions a 1dB degradation in CNR whilst subject
to interference, is a precursor to reduced or erratic performance (HPE etc…) in GPS receivers
• Rather than just testing to a mandated level of RFI, the work we’ve done highlights testing past
the 1dB degradation in CNR provides much deeper understanding of receiver behaviour
• Results also demonstrate stochastic nature of GNSS measurements – Too few measurements
in a Live Sky scenario (with even more variables) can give misleading results
• So far concentrated on single interference source – however LTE deployment plans means
there is a need for multiple interferer scenarios to test against…
15
Spirent Insights …and further work
• We are building up our “Big data” characterization/certification of receiver behaviour
including more parameters including:-
– RFI vs • TTFF (Cold, Warm, Hot)
• Reacquisition
• 1PPS and NTP/PTP accuracy
– Control Segment Errors
– Signal and Navigation data health status
– RAIM
– Multipath (Sim3D)
• RTK and MCMF
Join the GNSS Vulnerabilities group on Linked In to find out more
about GNSS jamming and spoofing the
Trust but Verify
https://www.spirent.com/PNT/Measuring-Resilience
Spirent® Communications, Inc. and its related company names, branding, product names and logos referenced herein,
and more specifically “Spirent” are either registered trademarks or pending registration within relevant national laws.
