SVN-49 Signal Anomaly Presented by Tom Stansell GPSW POC: Lt. Col. James Lake, Ph.D.

SVN-49 Signal Anomaly

Presented by Tom Stansell

GPSW POC: Lt. Col. James Lake, Ph.D.

2

DraftIntroduction

• GPS IIR-20, SVN-49 (PRN 01), carries an L5 Demonstration Payload – The L5 signal was not for operational use– The intent was to “bring L5 into use” for ITU purposes

• The demonstration payload made use of an Auxiliary Payload port on the spacecraft

• No impact on the L1 and L2 signals was intended or expected

• However, 2SOPS and Aerospace reported unusually high and elevation angle dependent Pseudo Range Residuals (PRR) from the monitor stations

3

DraftPseudorange Residuals

• Ionospheric refraction corrected pseudoranges• Relative to a “best fit” orbit determined early in the test program• Roughly a 4+ meter spread from 10 to 80 degrees

4

DraftGPS IIR Antenna Farm

L-Band Antenna Array with 12

Helical Antennas

5

DraftL-Band Antenna Element Locations

Outer Ring

Inner Ring

6

Draft

Outer Ring Inner Ring

L5 Filter162 In. Cable

Antenna Coupler Network

7

DraftJ1 and J2 Antenna Patterns at L1

0

2

4

6

8

10

12

14

16

18

20

22

24

0 10 20 30 40 50 60 70 80 90

Elevation Angle (deg)

L1

Ga

inl (

dB

)

J1 Signal Gain

J2 Signal Gain

Range Measurement Data

8

DraftReflected Relative to Direct L1 Signal Power

-40

-38

-36

-34

-32

-30

-28

-26

-24

-22

-20

-18

-16

-14

-12

0 10 20 30 40 50 60 70 80 90

Elevation Angle (deg)

L1

De

lay

ed

to

Dir

ec

t S

ign

al D

iffe

ren

ce

(d

B)

Reflected to Direct Signal Power Ratio

Range Measurement Data

9

DraftModel of L1 Signal Difference

-40

-38

-36

-34

-32

-30

-28

-26

-24

-22

-20

-18

-16

-14

-12

0 10 20 30 40 50 60 70 80 90

Elevation Angle (deg)

L1

De

lay

ed

to

Dir

ec

t S

ign

al D

iffe

ren

ce

(d

B)

Interpreted Power Difference

Delayed Signal Polarity Reversal

10

DraftEarly Minus Late Tracking

Effect of Half Voltage Amplitude Multipath Signal - C/A Code Scale

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5

Multipath Delay (C/A Code Chips)

Pseu

do

ran

ge E

rro

r (C

/A C

od

e C

hip

s)

C/A Code, Wide Correlator

C/A Code, 0.1 Narrow Correlator

293 MP Code Correlator

Max Code Error w/Opposite Polarity

Max Code Error w/Same Polarity

11

DraftIonospheric Refraction Calculation

2

2

2 2

2 2 2 21 2

1 2

2 2

1/ 2 1575.42 /1227.6 77 / 60

77 5929

60 3600

77 60 2329

( 77 60 ) /(77 60 )

2.55 1.55

2.55 1.55 2.984.... 3

L L

L L

L L

PR PR PR

PR PR PR

Ionospheric correction amplifies code noise

12

DraftMax Impact on L1 P(Y) Autocorrelation

0.0

0.2

0.4

0.6

0.8

1.0

1.2

-30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40

Pseudorange (meters)

L1

P(Y

) A

uto

co

rre

lati

on

Sum of the Two

P(Y) Autocorrelation

30 nsec Delay

13

DraftEarly Minus Late L1 P(Y) Code Track Error

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

-14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14

Pseudorange (meters)

Ea

rly

Min

us

La

te E

rro

r F

un

cti

on

Wide Dif

Narrow Dif

Narrow Sum

Wide Sum

1.62 m offset x 2.55 = 4.14 m error

Same Phase

Opposite Phase

Iono Correction Factor

14

DraftUseful Observation

• Receivers with early minus late correlators having similar spacing will have essentially the same tracking error for all signals at one frequency – e.g., L1 P(Y) and L1 C/A exhibit the same error

• However, different types of correlators and different correlator spacings very likely will produce different tracking errors– See next slide with figures from 13 July ’09 GPS World Article

15

DraftTracking Error with Different Correlators

L1 Typical Semi-Codeless Correlator

L2 Typical Semi-Codeless Correlator

L1 Using Multipath Mitigation w/ 20 nanosecond correlator spacing

L2 Using Multipath Mitigation w/ 20 nanosecond correlator spacing

Figures courtesy of GPS World, receivers are JAVAD GNSS Triumph receivers

16

DraftJ1 and J2 Antenna Patterns at L2

0

2

4

6

8

10

12

14

16

18

20

22

24

0 10 20 30 40 50 60 70 80 90

Elevation Angle (deg)

Ga

inl (

dB

)

L2 J1 Signal Gain

L2 J2 Signal Gain

Range Measurement Data

17

DraftModel of L2 Signal Difference

-50

-48

-46

-44

-42

-40

-38

-36

-34

-32

-30

-28

-26

-24

-22

-20

-18

-16

0 10 20 30 40 50 60 70 80 90

Elevation Angle (deg)

L2

De

lay

ed

to

Dir

ec

t S

ign

al D

iffe

ren

ce

(d

B)

Interpreted Power Difference

Delayed Signal Polarity Reversal

18

DraftEarly Minus Late L2 P(Y) Code Track Error

Same Phase

Opposite Phase

Iono Correction Factor

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

-14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14

Pseudorange (meters)

Ea

rly

Min

us

La

te E

rro

r F

un

cti

on

Wide Dif

Narrow Dif

Narrow Sum

Wide Sum

~.95 m offset x 1.55 = 1.48 m error~1.1 m offset x 1.55 = 1.69 m error

19

DraftRefraction Corrected Error Possibilities

• Assume the direct and reflected L1 signals are in phase so at zenith the L1 pseudorange is 1.62 m too long

• If the direct and reflected L2 signals are in quadrature, the L2 pseudorange error is negligible

• Therefore, the refraction corrected pseudorange error is (2.55 x 1.62 – 1.55 x 0) = 4.14 m

• If the direct and reflected L2 signals are in the same phase, the L2 pseudorange error is ~0.95 m

• Therefore, the refraction corrected pseudorange error is (2.55 x 1.62 – 1.55 x 0.95) = 2.66 m

• If the direct and reflected L2 signals are in opposite phase, the L2 pseudorange error is ~ -1.1 m

• Therefore, the refraction corrected pseudorange error is (2.55 x 1.62 – 1.55 x –1.1) = 5.84 m

20

DraftPseudorange Error Model

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

0 10 20 30 40 50 60 70 80 90

Elevation Angle (deg)

Ma

x P

se

ud

ora

ng

e E

rro

r (m

)

Iono w/ L2 direct and delayed opposed

L2

L1Iono w/ L2 direct and delayed aligned

Iono w/ L2 direct and delayed in quadrature

21

DraftCurves and Residuals Overlay

22

DraftA Partial Fix

• In order to reduce the elevation-dependent tracking residuals, 2SOPS has experimented with placing the antenna phase center about 152 meters above the satellite rather than slightly below as normal– (How can you fix a 4-5 meter problem with a 152 meter solution?)

• The Kalman filter then provides orbit and clock parameters which best fit the tracking data– The key parameter is clock offset

• Over the next few weeks, different values will be tried and transmitted in the NAV messages

23

DraftRaise the Orbit, Offset the Clock

• If Rs effective = Rs +

• The impact on pseudorange is cos(b)

• The following plot shows the effect of = 152.586 m with a clock offset of 496.2 nsec (148.754 m)

Rs =26578

Re =6378

Xr = Re Sin(a)Yr = Re cos(a)

a

bb = atan(Xr/(Rs - Yr))

El

Sin(90+El)/Rs = Sin(b)/Re

Cos(El)/Rs = Sin(b)/Re

El = acos(Rs Sin(b)/Re)

Rs ~ 26,560,601 meters

Re ~ 6,378,137 meters at equator

24

DraftNet Compensation

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0 10 20 30 40 50 60 70 80 90

Elevation Angle (deg)

Ps

eu

do

ran

ge

Co

rre

cti

on

(m

)

- Antenna Phase Center 152.586 m above SV- Clock offset 496.2 nsec or 148.754 m

25

DraftUncompensated Residuals

26

DraftCompensated Residuals

27

DraftCompare Compensation with Error Models

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

0 10 20 30 40 50 60 70 80 90

Elevation Angle (deg)

Ma

x P

se

ud

ora

ng

e E

rro

r (m

)

Iono w/ L2 direct and delayed opposed

L2

L1Iono w/ L2 direct and delayed aligned

Iono w/ L2 direct and delayed in quadrature

Orbit and Clock Compensation

28

DraftSpirent Simulation

• Note for organizations with a Spirent simulator– Spirent is preparing a scenario, based on these models, to simulate

the SVN-49 problem and enable laboratory testing– The scenario provides normal L1 and L2 signals plus a delayed

signal with the proper relative amplitude and phase relationships as a function of elevation angle in accordance with these models

– Several parameters can be modified by the operator– The scenario will available directly from Spirent by request