3/14/2025 LSG2202 1

GPS ERROR SOURCES

3/14/2025 LSG2202 2

GPS ERROR SOURCES

• Errors are differences between the actual value of a
quantity and the measured value.
• Measurement errors are often categorised as noise or
bias
➢Systematic errors (biases) –tend to persist overtime
but can usually be modelled mathematically or
eliminated using appropriate techniques

➢Random errors (noise) – quickly varying error that

averages out to zero over a short time interval
depending on the smoothing time of a receiver. Usually
unavoidable
3/14/2025 LSG2202 3

Sources of Error
1. Satellite
▪ Orbit
▪ Clock

2. Propagation medium – the atmosphere

▪ Ionosphere
▪ Troposphere

3. Receiver
▪ Clock
▪ Multipath
▪ Noise
▪ Antenna phase centre
3/14/2025 LSG2202 4

Sources of Error
 3/14/2025 LSG2202 5

Satellite Orbit Error

• Satellite positions are a function of time based on previous GNSS
observations at the ground control stations
• Modelling the forces acting on the satellites is not perfect --- errors in
the estimated satellite positions
• Orbit errors are therefore the difference between the predicted
and actual position of a satellite
• An orbit error for a particular satellite is identical to all GNSS users
worldwide
• However, users see the same satellite at different view angles ---
effect of the orbit error on the computed positions is different for the
different users
3/14/2025 LSG2202 6

Satellite Orbit Error

 3/14/2025 LSG2202 7

Satellite Orbit Error

Methods for minimising the effect of the orbit errors
1. Relative positioning i.e. combining the measurements of 2 receivers
tracking the same satellite can also minimize the effect of orbit errors
essentially removing it totally for short baselines.

2. For applications that require very high accuracy, the effect of the
orbit errors will be minimised by using the precise ephemeris instead
of the broadcast ephemeris
• Precise ephemerides are available from a number of sources, including
the International GNSS Service (IGS), the National Geodetic Survey
(NGS), and the Center for Orbit Determination in Europe (CODE)
3/14/2025 LSG2202 8

Satellite Clock Error

 3/14/2025 LSG2202 9

The Ionosphere
• First part of the atmosphere -50
km to 1000 km above the
earth’s surface
• A shell of negative electrons and
positively charged atoms and
molecules
• Charged particles exist due to
the impact of the sun’s ultra-
violet radiation
• Free electrons interact with the
GPS signals to affect signal
propagation
 3/14/2025 LSG2202 10

The Ionosphere
• Ionosphere is a dispersive medium i.e. bends the GPS signals and
changes its speed as it passes through the layers of the
ionosphere
• Bending causes a negligible range error as long as the SV elevation
angle is greater than 5˚
• The change in the propagation speed is the problem as it causes a
significant range error (Ionosphere is the largest single source of
error in GPS positioning)
• The ionosphere speeds up the propagation of the carrier phase
beyond the speed of light while it slows down the PRN code and
the satellite message by the same amount.
• The range will be too short if measured by carrier phase and too
long if measured by the code compared to the actual range
 3/14/2025 LSG2202 11

The Ionosphere
• The ionospheric delay is proportional to number of free
electrons along the GPS signal path
• Number of free electrons-Total Electron Content (TEC) i.e. a
measure of the number of free electrons in a column through
the ionosphere with a cross-sectional area of 1m².
• The effect of the ionosphere on the C/A-code can vary from 1 m
to 100 m depending on:
➢time of day (electron density maximum in early afternoon and
minimum at midnight)
➢Season (time of year-worse in winter than summer)
➢receiver location
➢viewing direction
➢solar activity
➢state of the earth’s magnetic field
 3/14/2025 LSG2202 12

The Ionosphere
Mitigation of the Ionospheric effects
• Option 1— Utilise the 8 parameters contained in the
satellite message to correct single receiver
observations. The correction is based on the model of
Klobuchar and can eliminate about 50% of the
ionospheric error at mid-latitudes

• Option 2— Do relative positioning (code or carrier

phase based) to eliminate common ionospheric errors.
This only works for “short” baselines (<20 km
approx.)
 3/14/2025 LSG2202 13

The Ionosphere
Mitigation of the Ionospheric effects
• Option 3— Use dual frequency data to account for 1st
order ionospheric effects.
• dispersive nature of the ionosphere, means that the
ionospheric delay is frequency dependent

• The lower the frequency the greater the delay

• The 28% separation between the L1 and L2 frequencies

facilitates the estimation of the ionospheric delay meaning
that by tracking both carriers, a dual-frequency receiver has
the facility of modelling and removing not all but a significant
portion of the ionospheric effect.
 3/14/2025 LSG2202 14

The troposphere
• Troposphere - earth’s surface to a
height of about 40 km

• Troposphere -neutral atmosphere

because it is non-dispersive for radio
waves with frequencies ≤15 GHz

• For GPS signals, tropospheric

refraction is independent of
frequency

• Modelling tropospheric delay by dual

frequency methods is therefore not
possible

• Troposphericdelay-a function of
temperature, pressure and relative
humidity
 3/14/2025 LSG2202 15

The troposphere
Two components:
1. Dry component (hydrostatic delay) -dry part of the
atmosphere-about 90% of the total tropospheric delay
2. Wet component - water vapour-the remaining 10% of the
total tropospheric delay. However variable in time and
space.

• Tropospheric delay depends on the distance travelled by

the signal through the neutral atmosphere and so is a
function of zenith angle
3/14/2025 LSG2202 16

The troposphere
Mitigation of the Ionospheric effects
Tropospheric error may be accounted for by:
1. A priori mathematical modelling (using
meteorological data)
▪Saastamoinen Model
▪Hopfield Model
▪Mapping Functions – NMF, IMF, VMF
2. Zenith path delay estimation (complex)
3. Differential positioning (limited to short baselines)
3/14/2025 LSG2202 17

Receiver multipath

• Multipath errors occur when

a satellite signal arrives at a
receiver via more than one
path (interference)
• Multipath is caused by
reflection of the satellite
signal off surfaces close to
the receiver
3/14/2025 LSG2202 18

Receiver multipath

• The multipath error is frequency dependent, affecting

code measurements more than carrier phase
measurements
• C/A-code pseudo-ranges can be in error by 10-20 m
(and up to 100 m) due to multipath errors
• The maximum error in a carrier phase measurement
occurs when the reflected signal is shifted 90° ( λ/4)
relative to the direct signal. This introduces a range
error of about 5 cm
3/14/2025 LSG2202 19

Receiver multipath
Ways of minimising multipath
1. Careful site selection- best way
▪Avoid sites near buildings, fences, water, cars, power lines etc...

2. Use of ground planes, or Choke Ring Antennae

3/14/2025 LSG2202 20

Receiver multipath
Ways of minimising multipath
3) Digital filtering of the observations

4) Avoiding signals from low elevation satellites -more susceptible

to multipath (cut-off angle)

5) Increasing observation time

3/14/2025 LSG2202 21

Note on Satellite Elevation

• Satellite elevation is crucial factor in the mitigation of
ionospheric effects, tropospheric effects and multipath.
• Signals from low-elevation satellites will have longer
distance to travel through the atmosphere than those at
high elevations. These will experience greater ionospheric
and tropospheric delays.
• To overcome these problems, signals from low-elevation
satellites are generally not used by inserting a cut-off or
mask angle in the receiver software to exclude them from
any processing- vertical angle of 5° and above the
observer’s horizon is used.
3/14/2025 LSG2202 22

Receiver clock offset

• GPS receivers use much less accurate clocks than
the satellites
• Receiver clock error much larger than satellite
clock error

Minimizing
1. Relative positioning – differencing between satellites
2. Treated as additional unknown parameter in the
estimation process
3. Precise external clocks can be used instead of the
internal clocks
Phase Centre Offset & Variation
• This error is the result of non-
coincidence between the geometric
centre and the phase (measurement)
centre of an antenna.
• Must differentiate between phase
centre offset and phase centre
variation
• For any antenna, the 3D phase centre
offset is constant, but variation
depends on satellite geometry
(azimuth /elevation)
• A dual frequency antenna has different
phase centres for L1 and L2
Phase Centre Offset & Variation Cont’d

Mitigation
• Phase centre offset and variation can be eliminated by parallel
antenna orientation if antennae are identical
• If different antennae are used, antenna calibration models must
be applied
• Calibration data for a wide range of antennae are available from
the NGS antenna calibration web-site
https://www.ngs.noaa.gov/ANTCAL/
3/14/2025 LSG2202 25

Satellite geometry
• The distribution of the satellites above an observer’s horizon
has a direct bearing on the quality of the position derived from
them

• DOP factors (Dilution of Precision)

• Numerical representation of satellite geometry
• Dependent on the locations of satellites that are visible to the receiver

• A low DOP – good- more precise the position/time computed

• A high DOP -bad
3/14/2025 LSG2202 26

Satellite geometry
3/14/2025 LSG2202 27

Satellite geometry
3/14/2025 LSG2202 28

User Equivalent Range Error (UERE)

In combination with a DOP factor, the UERE allows for an estimation
of the achievable point positioning accuracy

Error Source SPS PPS

Satellite orbit 2.1 2.1

Satellite clock 2.1 2.1

Ionosphere 4.0 1.2

Troposphere 0.7 0.7

Multipath 1.4 1.4

Phase centre 0.0 0.0

Noise 3.0 0.3

UERE 6.0 3.6