GPS Receiver Technology

1
GPS Receiver Technology

• Stuart Riley
• October 2005

2
Overview

• New Technologies Performance
• GPS Modernization Testing
• Multipath
• RTK Performance

3
Wideband Antenna L1/L2/L5

• Based on Proven Zephyr Technology

4
Maxwell Evolution
Power W
Maxwell V
Maxwell
Maxwell III
Maxwell IV
5
RF Improvements

• IF RF SAW Filters
• New flexible RF ASIC

6
Increased Integration
L1/L2/L5 RF Bands
L1/L2 RF Bands
7
Trimble R8 GNSS Features

• L1/L2/L5 Signal Tracking
• Negligible power penalty with L5 tracking
• Bluetooth for cable less operation
• Removable Rechargeable Battery
• External Power Input
• New Zephyr Derivative Antenna Element
• Updated design to include L5
• New CPU
• Increased onboard data logging
• Various Radio Modem Options

8
Low Elevation Tracking System Performance

• UNAVCO TEQC software observed to expected

9
High Elevation Tracking System Performance

• UNAVCO TEQC software observed to expected

10
Repeatability of Zephyr Antenna PCV

• 10 Samples
• lt0.2mm L1
• lt0.4mm L2

11
Zephyr Code Multipath Rejection

• MP1 and MP2 statistics from TEQC software

12
Civil L2

• Two codes, one with and one without message data
• Serves both commercial and consumer interests
• Code characteristics
• Codes longer than C/A to minimize cross
  correlation
• Time multiplexed, alternate code chips
• Combined chip rate is 1.023 MHz
• Average clock rate for each code is 511.5 kHz
• 20 msec code with 10,230 chips
• 1.5 sec code with 767,250 chips (gt 45 dB
  correlation protection)
• First Satellite Transmission 2005-10-21
• PRN 17 Currently Unhealthy

13
L2C Signal Strength
14
L2C Productivity Advantage

• 84 36 degree Elevation

15
Civil L5

• I-channel
• 50Hz data
• Encoded with FEC 100bps
• Encoded with 1KHz 10-symbol Neuman-Hoffman Code
• Modulated with X(I) code at 10.23MHz
• Modulated onto the carrier
• Q-channel
• No data message
• Encoded with 1KHz 20-symbol Neuman-Hoffman Code
• Modulated with X(Q) code at 10.23MHz
• Modulated onto the carrier

16
L2C/L5 Testing

• JPO L2-C / L5 Modulator Tests
• L1/L2/L5 Simulator

17
The Multipath Problem

• Multipath occurs when multiple copies of a
  satellite signal arrive at the antenna from
  different paths

SV 8
SV 8
SV 8
18
How Multipath Occurs
SV 8

• Multiple copies are created when the satellite
  signal reflects off of objects near the antenna

SV 8
SV 8
SV 8
19
Three Satellite 8s ???
SV 8
SV 8
SV 8

• Three signals confuse the receivers signal
  correlation process causing range errors.

SV 8
SV 8
SV 8
20
Code Tracking Basics
Receiver PN Code
Input base-band samples

• Channel multiplies the reference PN with the
  input samples
• Result goes into an integrate-and-dump filter
• Error phase between reference and input data,
  error filtered
• and applied to hardware.
• Multipath adds a second (or more) image of the
  PN code to
• input samples.

21
Code Tracking Discriminator
Punctual (P)
Late (L)
Early (E)
Tracking Error Chips
-1
1
22
Code Tracking Discriminator

• Shape is rounded in a real receiver
• Any unpredicted distortion may cause a code
  tracking error
• Error is filtered and used to maintain lock and
  provide the pseudorange observable.

23
Code Tracking ACF
1.5
1
0.5
ACF
0
-0.5
-4
-2
0
2
4
1.5
1
0.5
ACF
0
-0.5
-4
-2
0
2
4
Delay Chips
24
Code Tracking ACF

• Transitions from constructive to destructive
  interference
• L1 Carrier Wavelength 19cm
• Rate of change determined by environment
• High rate of change multipath is filtered by code
  smoothing

25
Generic Narrow Correlator Error Envelope
Multipath Envelope
8
6
4
2
0
Multipath Error Metres
-2
-4
-6
-8
0
50
100
150
200
250
300
350
Multipath Delay Metres
26
DGPS Comparative Analysis

• 24-Hour Data Set
• Short base-line on flat roof
• Trimble Receiver with Everest Installed

27
Stationary Range Comparison

• Stat. Rng. Pseudorange - Accumulated delta
  range
• Ionospheric divergence modeled
• Provides pseudorange noise (including multipath)
  comparison.

28
Stationary Range Example
Trimble Everest 1-Sigma 18.16cm
2
1
Stat. Rng. m
0
-1
-2
0
1000
2000
3000
4000
X 1-Sigma 33.51cm
2
1
0
Stat. Rng. m
-1
-2
0
1000
2000
3000
4000
Time Seconds
29
Real-TimeDGPS Tests
Without Everest
Everest
30
RTK Performance

• New RTK engine tested in real-time and with
  regression data
• Significant improvements in time to initialize
  while maintaining solution integrity
• New RTK and other receiver enhancements yield

31
Questions ?