Simulation vs. Real World Testing: How to undertake controlled testing of your GNSS receiver design

Simulation Versus
Real World Testing
How to undertake controlled testing of your GNSS receiver design

Should global navigation satellite system GNSS receivers ONLY be
tested using real world signals to guarantee their proper operation?

No
That would be wrong and impractical

Here’s why...

SPIRENT eBook Page 2

Here’s why
Simulation, by definition, reproduces the signals and effects a
receiver sees in the real world, but under the controlled conditions of
the laboratory, making GNSS receiver testing a far more consistent
and reproducible exercise.


Importantly, the signals from the simulator are exact known
quantities… right down to the bit level. And, by providing the
ability to test the different performance parameters of the receiver
individually (or in concert), the simulated environment will provide
much clearer insight into the true performance of the receiver.


The Nine Key Tests
There are nine key tests that together provide the fundamental
characterisation of performance of any GNSS receiver:

1. Cold-start time to first fix
2. Warm-start time to first fix
3. Hot-start time to first fix
4. Acquisition sensitivity
5. Tracking sensitivity
6. Reacquisition time
7. Static navigation accuracy
8. Dynamic navigation accuracy
9. Radio frequency interference


Some tests may not be as critical as others, depending on
the intended application of the receiver. But between them,
these nine tests cover all the important aspects of receiver
performance that will impact the end-user experience.


Test 1: Cold-start time to first fix (TTFF)
This is one of the great tests of a GNSS receiver because it will be the
first thing that a user notices. Time to First Fix is always an important
metric, and the “cold-start” version is just that — the receiver is starting
from scratch, with no memory of any previous reading.
The time is unknown, the current almanac and ephemeris are unknown,
and (obviously) the current position is unknown.

SPIRENT eBook

It is also a test that is far better performed with a simulator,
because, the sure way of measuring this quantity is to run
a series of tests on each receiver and take an average time,
with each test based on a completely new location – several
thousands of kilometres from the previous one.

Try running that test using real-world satellite signals
when time cannot be stopped or re-wound!


Test 2: Warm-start time to first fix
The second test is similar to the first, but the difference is
important. For the so-called “warm start”, the time and almanac
are retained within the receiver’s memory. However, the ephemeris
data are either unknown or out of date and the position is within
100km of the last fix.


And while single measurements can be performed equally as well in
the real world, the added control of using a simulator in the laboratory
(and absence of outside influences) allows the test to be performed
with total certainty. Using a simulator, it’s also readily possible to take
multiple measurements and average the results. And when you have
altered your design or set-up, you can quantify the improvement by re-
running exactly the same tests with exactly the same conditions.


Test 3: Hot-start time to first fix
Although the hot-start TTFF is the least arduous of the time to first
fix measurements for the receiver, in many ways it’s probably the
most important, as this will be the performance that the end user
will experience most often.


In this scenario, the receiver has full data on time, the almanac and the
ephemeris, and the position is within 100km of the last fix. All that is
required is for the receiver to collect the full navigation message from
the simulator. As with other TTFF tests, and due to the importance of the
measurement, it is advisable to run the test several times with different
satellite geometries to calculate an average value for the TTFF.


Test 4: Acquisition sensitivity
The sensitivity of any GNSS receiver is key to its performance,
and acquisition sensitivity is the first for these important
measurements as it defines the minimum received power
level at which the receiver can obtain a fix.

SPIRENT eBook

This is another test where the simulator is an essential tool. It is only
through the ability to control the power output from the simulator
(on individual satellites, or all at once) that an accurate measure of
acquisition sensitivity can be obtained.

Ideally, the simulator should be capable of very fine power control
to within 0.1dB to obtain the most accurate possible measure of
acquisition sensitivity.


Test 5: Tracking sensitivity
As with acquisition sensitivity, the fine control of power levels is
essential in determining tracking sensitivity – the minimum power
level at which the receiver can maintain lock. Crucially, it is tracking
sensitivity measurements that will highlight the errors inherent in the
design of the receiver’s PLL-based tracking loops. These include phase
error, dynamic stress error and thermal noise.


The test itself is relatively easy: with the receiver locked
on to the simulator’s output, simply lower the simulator
power output until the lock is lost. Multiple repeats of
the test with different satellite geometries will ensure
that an accurate average measure is recorded.


Test 6: Reacquisition time
Reacquisition time is a particularly important measurement for
vehicle-based receivers, which will inevitably lose satellite signals
when travelling through tunnels or even under bridges. For example,
the end user will not be impressed if the receiver misses a turn
instruction because it has not reacquired the signal after passing
such an obstruction.

SPIRENT eBook

Again, the simulator allows total control over the test, reducing the
signals from each satellite by at least 60dB to ensure that the receiver
loses complete lock, raising it again to normal power and measuring the
time taken to reacquire the lock.


Test 7: Static navigation accuracy
This is in many ways the most difficult test to predict, as there are
so many different factors – both internal and external – that can
affect a receiver’s performance. So this is another case where the
controlled environment of the laboratory is essential to remove
external variables such as the effects of the ionosphere and
troposphere – or indeed include them, but in a quantified way.


A useful tip here is to simulate a static position of 0 degrees latitude,
0 degrees longitude and 0 metres elevation, as it will make it easy to
observe the receiver’s divergence from the simulated position.

Again, multiple measurements should be taken to allow for different
satellite positions and factors such as receiver thermal performance.
The resulting metric is typically quoted as a statistical average of the
many performance measurements.


Test 8: Dynamic navigation accuracy
Particularly important for vehicle-mounted receivers, dynamic
navigation accuracy involves taking a series of measurements
while the receiver is moving in one, two or three axes. While such
measurements could theoretically be taken reproducibly on a test
track, the simulator again has a trick up its sleeve that inevitably
leads to improved measurement accuracy.


The simulator control software has the ability to simulate the relative
motion of the receiver and satellites. And with a high dynamic
performance simulator, this means that virtually all types of vehicle
motion profiles can be simulated, with high fidelity even with the most
extreme maneouvers.


Test 9: Radio frequency interference
Because GNSS receivers are such sensitive instruments, it is almost
inevitable that they are susceptible to radio frequency interference –
most of which will be accidentally generated. However, there are also
instances where a jamming signal might be deliberately broadcast in
order to lock out a navigation system.


There are many commercial interference simulators on the market
that can be used to obtain a measure of a receiver’s susceptibility to
any given frequency of RFI. However, by using a coherent interference
source that is directly coupled to the GNSS simulator and dynamically
controlled by the same system software, far greater insight into a
receiver’s performance can be obtained, allowing designers to take
appropriate filtering measures to improve their products’ performance.


To conclude
The use of a multichannel RF constellation simulator in testing the
performance of a GNSS receiver has many benefits.

Indeed, many of the parameters that are key to the performance of a
receiver simply cannot be reliably tested in the real world – in some
cases due to simple practicality, in others because of external variables
that render test results unreliable.


It is only by rigorous simulator-based testing, under the controlled
conditions of the test laboratory, that the nine key performance
indicators of any GNSS receiver can be determined with absolute
certainty and full repeatability. Armed with these results, GNSS
receiver developers can perfect their designs, leading to improved
products that exceed users’ expectations.


Spirent GNSS Simulators
Spirent is the industry leader for GNSS simulator products. Spirent
offers several different models of GNSS simulators that support a
variety of different applications and cover the full spectrum of civilian
and military GNSS testing needs. Spirent products range from basic
single-channel simulators, suitable for simple production testing,
through multi-channel, multi-constellation simulators, suitable for the
most demanding research and engineering applications.


For more comprehensive testing, Spirent also offers products that
simulate additional system elements simultaneously with the GNSS
constellation signals, such as inertial sensors, various automotive
sensors, Assisted GPS (A-GPS) + Assisted GLONASS (A-GLONASS)
data, SBAS and GBAS augmentation system signals, interference
signals and Wi-Fi Positioning.

Spirent GSS8000 Spirent GSS6700 Spirent GSS6300
Multi-GNSS Constellation Multi-GNSS Constellation Multi-GNSS Signal
Simulator system generator


We hope you found this Spirent Simulation vs Real World
Testing E-Book of interest.
We are continually adding new content to our website
on a regular basis. Bookmark this link:
www.spirent.com/positioning

Visit the Spirent GNSS Blog, there are currently
more than 90 posts with 2 to 3 new posts added
per week. Catch up on what’s new:
www.spirent.com/Blog/Positioning

Need more information?
gnss-solutions@spirent.com

Share?

Facebook LinkedIn Twitter Technorati Google Buzz Digg Delicious Reddit Stumbleupon MCD00170 Issue 1-02

Spirent Spirent Federal Systems Got a smartphone?
+44 1803 546325 +1 714 692 6565 Scan the QR
globalsales@spirent.com info@spirentfederal.com Code for more
www.spirent.com/positioning www.spirentfederal.com information

Simulation vs. Real World Testing: How to undertake controlled testing of your GNSS receiver design

More Related Content

What's hot

Similar to Simulation vs. Real World Testing: How to undertake controlled testing of your GNSS receiver design

Recently uploaded

Simulation vs. Real World Testing: How to undertake controlled testing of your GNSS receiver design