Some time ago I did a few experiments about pushing 2kbaud 8PSK and differential 8PSK through the QO-100 NB transponder. I didn’t develop these experiments further into a complete modem, but in part they served as inspiration to Kurt Moraw DJ0ABR, who has now made a QO-100 Highspeed Multimedia Modem application that uses up to 2.4 kbaud 8PSK to send image, files and digital voice. Motivated by this, I have decided to pick up these experiments again and try to up the game by cramming as much bits per second as possible into a 2.7 kHz SSB channel.
Now I have a definition for the modem waveform, and an implementation in GNU Radio of the modulation, synchronization and demodulation that is working quite well both on simulation and in over-the-air tests on the QO-100 NB transponder. The next step would be to choose or design an appropriate FEC for error-free copy.
In this post I give an overview of the design choices for the modem, and present the GNU Radio implementation, which is available in gr-qo100_modem.
A few days ago, the spring eclipse season for Es’hail 2 finished. I’ve been recording the frequency of the NB transponder BPSK beacon almost 24/7 since March 9 for this eclipse season. In the frequency data, we can see that, as the spacecraft enters the Earth shadow, there is a drop in the local oscillator frequency of the transponder. This is caused by a temperature change in the on-board frequency reference. When the satellite exits the Earth shadow again, the local oscillator frequency comes back up again.
The measurement setup I’ve used for this is the same that I used to measure the local oscillator “wiggles” a year ago. It is noteworthy that these wiggles have completely disappeared at some point later in 2020 or in the beginning of 2021. I can’t tell exactly when, since I haven’t been monitoring the beacon frequency (but other people may have been and could know this).
A Costas loop is used to lock to the BPSK beacon frequency and output phase measurements at a rate of 100 Hz. These are later processed in a Jupyter notebook to obtain frequency measurements with an averaging time of 10 seconds. Some very simple flagging of bad data (caused by PLL unlocks) is done by dropping points for which the derivative exceeds a certain threshold. This simple technique still leaves a few bad points undetected, but the main goal of it is to improve the quality of the plots.
The figure below shows the full time series of frequency measurements. Here we can see the daily sinusoidal Doppler pattern, and long term effects both in the orbit and in the local oscillator frequency.
If we plot all the days on top of each other, we get the following. The effect of the eclipse can be clearly seen between 22:00 and 23:00 UTC.
By adding an artificial vertical offset to each of the traces, we can prevent them from lying on top of each other. We have coloured in orange the measurements taken when the satellite was in eclipse. The eclipse can be seen getting shorter towards mid-April and eventually disappearing.
We see that the frequency drop starts exactly as soon as the eclipse starts. In many days, the drop ends at the same time as the eclipse, but in other days the drop ends earlier and we can see that the orange curve starts to increase again near the end of the eclipse. This can be seen better in the next figure, which shows a zoom to the time interval when the eclipse happens, and doesn’t apply a vertical offset to each trace. I don’t have an explanation for this increase in frequency before the end of the eclipse.
The plots in this post have been done in this Jupyter notebook. The frequency measurements have been stored in this netCDF4 file, which can be loaded with xarray.
Some months ago I published the analysis of a BPSK pulse radar waveform that Scott Tilley VE7TIL had received through the transponder of Meridian 8 at a downlink frequency of 994 MHz. Now Roland Proesch DF3LZ has analyzed the same recording that I used, finding some different signal parameters. This has made me review my analysis, and it turns out that I made a mistake in finding the symbol rate of the signal. This post is an updated analysis, correcting my mistake.
Back in 2016, I became interested in Othernet‘s satellite filecasting service, which back then was called Outernet. Outernet was a start up company that offered a free global satellite service with which some internet contents such as weather updates or Wikipedia pages were broadcast. The main goal of the company was making this content available in developing countries, remote locations, and in general, in places that couldn’t afford an internet connection. Outernet started being popular with Amateur radio operators, hobbyists and makers, who saw it as an interesting project to set up at home for fun.
Most of the Outernet’s software stack was open source, and their receivers were some kind of embedded Linux system. However, the key parts of the receiver, which were the demodulator and the implementation of the filecasting protocol, were closed source and the protocols and specs they used were not documented publicly. Unhappy with this, I wrote in their forums asking if these software pieces could be open sourced. After receiving a negative reply, I decided to try to reverse-engineer all this, with the goal of writing some documentation and producing an open-source alternative implementation.
This is hardly news any longer. Since a few weeks ago, some people have been decoding the S-band telemetry from the Falcon-9 upper stage, which includes live video of the exterior of the spacecraft and the liquid oxygen tanks interior. This started with the work of r00t.cz, @aang254 and others, who around the second week of March managed to decode video from the telemetry for the first time. r00t.cz has published some information about the telemetry, @aang254 has added a decoder to SatDump, and Alexandre Rouma is adding a decoder to SDR++. In fact, the whole exercise of decoding the video has become quite popular, and more and more people are contributing to decode the latest launches.
On 2021-03-24 there was a launch of 60 Starlink satellites from Cape Canaveral, and Iban Cardona EB3FRN sent me the IQ recording he did on the first orbit, some 18 minutes after the launch. I decided to make a GNU Radio decoder flowgraph for this, since even though there are already several software decoders, I haven’t seen anyone using GNU Radio. I figured out that I could easily put together a flowgraph using some blocks from gr-satellites. This would make a useful and interesting example of using GNU Radio to decode digital signals.
In my last post about Tianwen-1, I explained how on 2021-02-23 the spacecraft would enter an orbit with a period of 2 Mars sidereal days. This would give a repeating ground track with periapsis passages over the intended landing site in Utopia Planitia. Almost one month has passed since then and AMSAT-DL has continued to receive telemetry state vectors every day with the Bochum 20 meter antenna. This data allows us to study the orbit in detail, including orbit perturbations and any station-keeping manoeuvres that are done to maintain the orbit. This post is my first analysis of the current orbit.
Back in 2019, I took advantage of the autumn sun outage season of Es’hail 2 to make some observations as the sun passed in front of the fixed 1.2 metre offset dish I have to receive the QO-100 transponders. Using the data from those observations, I estimated the gain of the dish and the system noise. A few weeks ago, I have repeated this kind of measurements in the spring sun outage season this year. This post is a summary of the results.
In the weekend experiments that we are doing with the GNU Radio community at Allen Telescope Array we usually have access to some three antennas from the array, since the rest are usually busy doing science (perhaps hunting FRBs). This is more than enough for most of the experiments we do. In fact, we only have two N32x USRPs, so typically we can only use two antennas simultaneously.
However, for doing interferometry, and specially for imaging, the more antennas the better, since the number of baselines scales with the square of the number of antennas. To allow us to do some interferometric imaging experiments that are not possible with the few antennas we normally use, we arranged with the telescope staff to have a day where we could access a larger number of antennas.
After preparing the observations and our software so that everything would run as smoothly as possible, on 2021-02-21 we had a 18 hour slot where we had access to 12 antennas. The sources we observed where Cassiopeia A and Cygnus A, as well as several compact calibrators. After some calibration and imaging work in CASA, we have produced good images of these two sources.
Many thanks to all the telescope staff, specially to Wael Farah, for their help in planning together with us this experiment and getting everything ready. Also thanks to the GNU Radio team at ATA, specially Paul Boven, with whom I’ve worked side by side for this project.
This post is a long report of the experiment set up, the software stack, and the results. All the data and software is linked below.
Last Saturday 2021-02-20 at 11:46:42 UTC Tianwen-1 passed the periapsis of its elliptical polar orbit at Mars and made a retrograde burn to reduce its apoapsis radius. The trajectory planning of the spacecraft can be seen in its Wikipedia page: the spacecraft first arrived into a low inclination elliptical orbit by making a Mars orbit insertion at periapsis, then coasted to apoapsis, where it performed a plane change, and then it arrived at periapsis, performing the manoeuvre described in this post.
Over the next few days the spacecraft should move into a reconnaissance orbit, which is given in Wikipedia to be a 265 x 60000 km orbit (having a period of 2 days) with an inclination of 86.9 degrees. However, the last burn hasn’t lowered the apoapsis that much. The current orbit is approximately 280 x 84600 km (3.45 day period) with an inclination of 87.7 degrees. A possible reason for using the current orbit, which has been described as a phasing orbit, will be explained in this post after reviewing the data we have about the burn.
This post has been delayed by several months, as some other things (like Chang’e 5) kept getting in the way. As part of the GNU Radio activities in Allen Telescope Array, on 14 November 2020 we tried to detect the X-band signal of Voyager-1, which at that time was at a distance of 151.72 au (22697 millions of km) from Earth. After analysing the recorded IQ data to carefully correct for Doppler and stack up all the signal power, I published in Twitter the news that the signal could clearly be seen in some of the recordings.
Since then, I have been intending to write a post explaining in detail the signal processing and publishing the recorded data. I must add that detecting Voyager-1 with ATA was a significant feat. Since November, we have attempted to detect Voyager-1 again on another occasion, using the same signal processing pipeline, without any luck. Since in the optimal conditions the signal is already very weak, it has to be ensured that all the equipment is working properly. Problems are difficult to debug, because any issue will typically impede successful detection, without giving an indication of what went wrong.
I have published the IQ recordings of this observation in the following datasets in Zenodo:
