performance assessment of a gps network rtk service

Performance Assessment of a GPS Network RTK Service


J. Aponte, X. Meng, A. Dodson, T. Moore, C. HillIESSG, The University of Nottingham, UK

M. BurbidgeLeica Geosystems (UK)


Content

• Objectives• Leica SmartNet• Methodology• Quality Measures

– Accuracy and Precision– Availability– Integrity– Repeatability– DOP

• Conclusions• Future Work• Acknowledgements


Objectives

• Assess performance of SmartNet service from end user’s point of view, based on:

– Accuracy

– Precision

– Integrity

– Availability

– Repeatability

– Number of satellites and DOP


Network RTK GPS Service: Leica SmartNet

• Partnership: Leica Geosystems and OSGB, OSI, and OSNI

• 142 RS– 16 RSs in Ireland.– 20 Leica RSs (8 GPS/GLONASS RSs).

• Operational from January 2006 in GB and July 2007 in Ireland.

• GPS receivers supporting RTCM formats 2.3/3.1

• Leica SpiderNet

• RSs: Leica 500/1200 receivers and Chokering Antennas.

NRTK GPS service for Great Britain and Ireland:


Assessment Methodology

AT502 DF Antenna

Antenna Signal Splitter

Leica Geosystems SR-530 GPS

Receiver

Leica Geosystems

GX-1200 GPS Receiver

Raw data logged in Receiver’s Memory

Card for Post-processing

Raw data logged in Receiver’s Memory

Card for Post-processing

Real Time NRTK solution saved in a Laptop for analysis

(NMEA GGA format)

RTCM 3.1 corrections received via

GPRS

Equipment Configuration

• NRTK Service tested “as it is”

• Static Tests: accuracy, precision, availability, integrity and repeatability

• Kinematic Tests: accuracy, availability and integrity

• TK1 covered a built-up area

• Influence number of GPS satellites and their geometry, age of the NRTK corrections (AoC)


Assessment Methodology

Static Tests Kinematic Tests


Quality Measures: Accuracy and Precision

2D RMSE

3D RMSE

ΔE ΔN ΔH (cm) (cm) E N H

TS1 -1.03 -3.27 4.36 3.43 5.55 0.13 0.50 0.62TS2 -0.98 -3.72 1.68 3.85 4.20 0.05 0.11 0.33TS3 -1.17 -2.97 3.61 3.19 4.82 1.24 2.04 3.72TS4.1 -1.82 -2.02 4.19 2.72 4.99 1.23 1.23 2.38TS4.2 -1.60 -3.07 3.40 3.46 4.85 2.80 2.66 5.40TS5.1 -1.12 -2.71 3.56 2.93 4.61 1.13 1.86 3.91TS5.2 -1.56 -2.91 4.34 3.30 5.45 1.14 2.01 3.59TK1 -1.05 -3.53 1.60 3.68 4.02 0.48 0.74 1.56TK2 -1.11 -3.52 2.35 3.69 4.38 0.54 0.90 1.69

Accuracy (cm)

Precision (+/- SD cm)Test

Accuracy and Precision Obtained during both Static and Kinematic Tests (NRTK Epochs). Accuracy averages:

0.70 to 1.27 cm Easting5.22 to 3.08 cm Northing4.55 to 3.23 cm Height

Precision averages:+/- 5.41 to 0.97 cm East+/- 7.37 to 1.34 cm North+/- 12.67 to 2.58 cm Height

Accuracy (error), how far the coordinates observed during testing were from the true values.

Precision, degree of closeness that repeated measurements displayed (Standard Deviation of the observations, 1 Sigma, about 68% of observations).



ENH Errors during TS4.1

-200

-100

0

100

200

-200

-100

0

100

200

0 1000 2000 3000 4000 5000 6000-200

-100

0

100

200

Time (sec)

Err

or

(cm

)

Easting

Northing

Height

All Epochs

-6

-4

-2

0

2

-6

-4

-2

0

2

0 1000 2000 3000 4000 5000 6000-5

0

5

10

15

Time (sec)

Err

or

(cm

)

Easting

Northing

Height

NRTK Epochs



0%

20%

40%

60%

80%

100%

ΔE (%) ΔN(%) ΔH(%) ΔE (%) ΔN(%) ΔH(%)

All Epochs NRTK Epochs

> 54 - 53 - 42 - 31 - 20 - 1

Class(cm)

Static Tests

0%

20%

40%

60%

80%

100%

ΔE (%) ΔN(%) ΔH(%) ΔE (%) ΔN(%) ΔH(%)

All Epochs NRTK Epochs

> 54 - 53 - 42 - 31 - 20 - 1

Class(cm)

Kinematic Tests

Percentage of Observations with Accuracy better or worst than 5 cm.


Quality Measures: Availability

Availability, percentage of observations in which a NRTK solution (integer ambiguities resolved) was obtained

NRTK Availability results

TestPossible Epochs

NRTK Epochs

Availability (%)

TS1 7200 6356 88.28

TS2 5400 5257 97.35

TS3 28800 25490 88.51

TS4.1 6600 4628 70.12

TS4.2 6600 6354 96.27

TS5.1 10800 9241 85.56

TS5.2 10800 10164 94.11

TK1 93599 50905 54.39

TK2 64799 40736 62.87

0%

20%

40%

60%

80%

100%

TS1 TS2 TS3 TS4.1 TS4.2 TS5.1 TS5.2 TK1 TK2

Missing Epochs (%) Lost Lock Epochs (%) Navigated Epochs (%)

DGPS Epochs (%) Availability (%)

Detailed percentage of observations performed under the different solution types during tests.


Quality Measures: Availability

TK1 2D RMSE Errors as Observed in the Covered Route

All Epochs

0 – 1 cm

>5 cm

1 – 5 cm

NRTK Epochs

Factors Affecting the Availability during the Kinematic Tests


Quality Measures: Integrity

Integrity, percentage of NRTK observations with accuracy better than 5 cm.

0%

20%

40%

60%

80%

100%

TS1 TS2 TS3 TS4.1 TS4.2 TS5.1 TS5.2 TK1 TK2

ΔE (%) ΔN(%) ΔH(%) ΔE (%) ΔN(%) ΔH(%)

NRTK Epochs All Epochs

Percentage of observations with errors better than 5 cm (Integrity).

-10

-5

0

5

10

-15

-10

-5

0

5

0 1000 2000 3000 4000 5000 6000-20

-10

0

10

20

Time (sec)E

rro

r (c

m)

Easting

Northing

Height

ENH Errors during TS4.2 (NRTK Epochs).


Quality Measures: Repeatability

Repeatability, similarity between the errors obtained during two different tests carried out under similar conditions (NRTK Epochs)

-10

-5

0

5

-10

-5

0

5

0 1000 2000 3000 4000 5000 6000-10

0

10

20

Time (sec)

Err

or

(cm

)

Easting

Northing

Height

Repeatability between TS4.1 (Light Colour) and TS4.2 (Dark Colour). 4331 common

epochs of a total of 6600 epochs.

-5

0

5

-10

-5

0

5

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000-10

0

10

20

Time (sec)

Err

or

(cm

)

Easting

Northing

Height

Repeatability between TS5.1 (Light Colour) and TS5.2 (Dark Colour). 8233 common

epochs of a total of 10800 epochs.


Positioning Quality Results N. SVs in view and DOP

Influence of N. SVs in view and their geometry over the positioning solution

Influence of the Number of GPS satellites in View over the Errors during TS4.2 (All Epochs).

Influence of the HDOP over the Errors during TS4.2 (All Epochs).

0 1000 2000 3000 4000 5000 6000-40

-30

-20

-10

0

10

20

30

40

Time (sec)

Err

or

(cm

)

0

1

2

3

4

5

6

7

8

9

Nu

mb

er

of

Sa

telli

tes

SVs Number

East North Height

0 1000 2000 3000 4000 5000 6000-40

-30

-20

-10

0

10

20

30

40

Time (sec)E

rro

r (c

m)

0

1

2

3

4

5

6

7

8

9

10

HD

OP

HDOP

East North Height


Conclusions

• SmartNet offers high centimetric accuracy (1.27 cm Easting, 3.08 cm Northing and 3.23 cm Height) for tests.

• It also achieves high precision (1-sigma) (+/- 0.97 cm East, +/-1.34 cm North and +/- 2.58 cm Height).

• Availability of the NRTK solution is the main drawback. Centimetric accuracy and precision only achieved during about 90% of the time in static tests and 58% during the kinematic tests.

• Lack of availability mainly due to external factors and not directly related to the NRTK systems:

– GPS signal disturbances (blockage).– Low number of GPS satellites in view and bad DOP.– Interruptions of the GPRS communication link.


Current and Future Work

• Investigating the impact of different commercial GSM/GPRS wireless services over the NRTK GPS availability.

• Examining the performance of the different NRTK concepts such as VRS and MAC.

• Study and development of NRTK GPS quality measures for service providers and final users.


Acknowledgements

• UK’s Engineering & Physical Sciences Research Council (EPSRC) for partially funding this project to the first author through its MTP scheme at the University of Nottingham.

• Special thanks to Leica Geosystems (UK) for providing, free of charge, the SmartNet connection used during this research.

• PhD studentship to the first author is jointly awarded by the EPSRC under its CASE PhD Studentship scheme and Leica Geosystems (UK).



J. Aponte, X. Meng, A. Dodson, T. Moore, C. HillIESSG, The University of Nottingham, UK

M. BurbidgeLeica Geosystems (UK)

performance assessment of a gps network rtk service

Documents

gps network rtk servicej

nrtk gps service

dopperformance assessment

accuracy averages

leica rss

leica smartnetpartnership

leica spidernetrss

height precision averages