a Virtual Robotix Network Team

Guy's I've been long interested in running an experiment using the GPS Front End form a radiosonde to get real-time speed and position data. Here is a teardown of one of these sondes

The reason I got so interested in these is the software program Sondemonitor With this software and a typical radio scanner you can process the data from these sonde's in realtime.

What I find really interesting is the simple GPS Front End used in these sonde's

There are several DIY GPS receivers that are extremely cool (and difficult) I have a great deal of respect for the guy's that where able to do this.

This one is a more modern version of the original done by

Notice the bulky front-ends on these DIY GPS look a lot like the tiny one in the radiosonde

What I find very interesting is the Sondemonitor software can give a realtime position and a bearing as well as output NEMA data about the Sonde.

 My goal here is to use the radiosonde front-end and using a raspberry pi or VR Nuron along with an algorithm similar to the Sondemonitor program to get realtime GPS position and speed from the "Doppler Measurements" that the sonde's collect.

I know that the Sonde's use SPI interface so I've ordered a Bus Pirate to help out along with the new 200Mhz DSO we just bought.

If there is any interested in helping with this project please comment!

Sondemonitor uses ephemeris data from the internet and the doppler measurements from the sonde's to get the fix and what we need is an algorithm to streamline the process.

Views: 2013

Replies to This Discussion

Wow that's very interesting project ... but for develop this kind of app is not possible to use SDR application ? I see that with this kind of new technology is possible to develop a lot of application .



Yes! It is indeed possible to use SDR and I have a TV dongle to use for experiments if anyone is interested in that type of experiment. BUT! It requires a great deal of processing to do realtime positioning in software with it. Right now it can be done with high-end video cards (GPU's) but we don't have that kind of processing power on our UAV's :)

I have another project to record the raw signal (not the raw GPS data) and post process it that can be done after the fact and get very precise data.

But this project is to use only the Doppler measurements provided by the radiosonde's front-end which is very much like a (Half-GPS) this allows a 2 Hz position fix if prior to launch you provide the software with ephemeris data (obtained from the internet) a description from the manufacture


The GPS receiver of GPS-radiosonde meas-
ures L1 carrier shift frequencies (=Dop-
pler’s) using a unique codeless detection
and measurement technique. The measured
Doppler shifts are a combination of satellite
and radiosonde motion and contain the
pertinent information for wind speed and
direction computation. Doppler frequency
measurements of up to 8 satellites are sent
to the ground equipment using a low
bandwidth (1200 baud) digital link. New
independent GPS Doppler measurements
from the GPS-sonde are provided every
0.5 s for up to 8 channels with 24-bit
The ground equipment computes the wind
speed and direction using the differential
GPS concept. Local carrier phase Doppler
shift measurements to satellites are ob-
tained from a high-quality 12-channel GPS
engine. The remote and local Doppler
measurements along with GPS time and
satellite ephemerides form the basic raw
data for wind computation.
New independent unfiltered velocity
components and filtered wind vectors are
provided with an update rate of 2 Hz. The
filtering is required for pendulum removal
which is the dominant feature at the 0.5-
second raw velocity vector level.
The windfinding accuracy is very good
even in the case where a mobile sounding
station is used. The slight degradation in
accuracy is mainly due to SA (Selective
 This is the description of the RS80 radiosonde the RS92 I have is similar except it has 12 channels and can be removed from the circut board before the ASIC that does the psudorange calculations and function the same as the RS80 in that respect except with 12 channels in an even smaller package :) and that may also allow for 3 Hz operation.
Why go to all the trouble? Well, I can purchase a Novtel GPS unit that can do 50 Hz updates and does not have COCOM limits (for use in spacecraft) But it cost $10,000 and another $1000 for the license.
What this would do would allow use to use the GPS system without the (secret) built in 4 G limit that causes loss of signal lock or the other COCOM limits of speed and altitude.
The only drawback is the 2 or 3 Hz update limit but for next to nothing in cost I can live with that.



© 2021   Created by Roberto Navoni.   Powered by

Badges  |  Report an Issue  |  Terms of Service