DIY spying on Satellites


When I was still a child, everything that happened in space interested me too much. Understanding how radio waves, coming from satellites such as television, telecommunications, meteorology, came to earth was surprising. At least, seen from the inexperience of my retinas.

Since always, and having a father radio amateur was simple for me to understand a little more about these pieces of tinplate that revolve around the planet. The technical literature on radio, electronics, telecommunications, and so on was abundant at home. Above all, the radio amateur magazines of the 80-90, where telecommunications were a rage. Many of them spoke about low-orbit satellites, and many others (from the mid-1990s) of the first Argentine satellite: LUSAT-1. I will not talk about this great satellite but believe me that reading those stories in my youth was something that marked my life, but knowing that I had the possibility of perhaps hearing that CW beacon or a carrier of two radio amateurs communicating.

Investigating, I found very valuable information from low-orbit meteorological satellites, known by their NOAA-xx identification, where XX is a number that identifies each satellite in particular. Currently we have NOAA-15, NOAA-18 and NOAA-19 in operation, and they travel the earth to an altitude of approximately 850KM with an orbital period of 100-105 minutes (around the world in less than two hours).

NOAA-19 in space

These satellites are equipped with various sensors and cameras that capture the world from above, through an analog transmission system called APT (Automatic Picture Transmission). Without entering into much technical detail about this technology, APT relies on transmitting two image channels (Video A / B), telemetry information and synchronization information. All this information is transmitted in 137Mhz, with 24khz of bandwidth at 4096 baud and 5 watts of power, something really very slow but impressive for its results.

All that information travels encoded as audio (as did the telephone modems) and a software is responsible for decoding that information, and making use of telemetry and synchronization information, achieve a perfect picture.

My first receptions

I started my first experiments in 2005, when I was 14-15 years old. Using a VHF receiver, almost always a Yaesu FT-2400 base or a Yaesu FT-51r handy with a vertical antenna, manufactured for 144Mhz and a desktop computer with a sound board, a Pentium III of 550Mhz.


“When will I hear the satellite? Can I predict it? How do I adjust the orbits?

To these questions, there was software that solved this problem. The still upbeat and cool Orbitron:

Orbitron is a free software that allows predicting the orbit of satellites using Kepler elements. These elements are nothing more than text files with calculated information from satellite orbits. All the time they are updated to correct the prediction, and that this one is the most exact possible. But … Where did you get the keplers? From BBS! (-:

Besides that also has a very good function, which is the automatic Doppler adjustment.

Being the satellite, an element that transmits information and moves along an orbit, the frequency received from an immovable place, varies up or down, as the element moves. Orbitron performs calculations based on the altitude and position of the satellite, to inform us the actual frequency of reception from our receiver.

All right! Thanks to Orbitron I can already calculate my orbits and I discovered that I can also adjust the frequency according to the Doppler that I am having at that moment.

When I listen to the satellite … What do I have to do? How do I decode that information? “

That was another big question. Having everything adjusted to the frequency of the satellite, knowing when and in what form reception will occur, what do I have to do next?

WXToIMG is the answer! WXToIMG is a shareware that allows you to decode the signal, the video channels, and achieve a real image of the planet, from the eyes of the satellite. In addition, it allows to make some “enhacements” to achieve better photos.

Here I am getting my first picture!

You’ll probably think … but what the fuck is that picture! But at my late fifteen, that was fascinating, incredible, from another planet. The nerves, the excitement, the lack of air, could not stop from jumping from the emotion. I spent a good amount of time so exalted until I could calm down and analyze what I was seeing. From now, the image was very bad, a lot of noise in the band, little signal from the satellite made the photo come out as ugly. It took me some months to finish understanding the flaws I had. In principle, the antenna I was using was not the right one. As a vertical antenna has no gain in the parts that most interest when working with satellites.

If we analyze the radiation pattern of a ground-plane antenna like I was counting then, we will see that it has almost no gain at high angles (where one more benefits from the passing of a satellite.

The receiver was also not suitable. NOAA satellites work at 137Mhz with a bandwidth of 24Khz, but most amateur radio equipment at that time has a maximum width of 15khz. So I was losing a lot of information because I did not have those extra 9Khz (4.5Khz on each side of the center frequency)

As a side note, you can see the shift of the telemetry (grayscale pictures to the sides of the image) due to the Doppler, since these amateur radio equipment has a step-size of 2.5khz minimum, Y Running 2.5khz would be moving me a lot on the frequency, risking me to lose even more information.

They spent many years trying to improve the setup I had but I did not get much better pictures than that. I think this was the best image I got:

In the image you can see the south of Argentina, the mountain range, Puerto Deseado, part of Chile.

The new

A few years ago, SDR receivers became fashionable. An SDR receiver, as the name implies, allows software-defined radio. This means that one has a wide bandwidth receiver, all conceivable modes and a wide range of frequencies for running. Thus, a person wrote a bookstore to use a digital television receiver, based on a Realtek RTL2832U as an SDR receiver. That way for 5-10 dollars we can buy a receiver like that and use it for our experiments.

These receivers operate from approximately 1 to 1800 Mhz, depending on the driver being used. The receiver is extremely responsive and allows working in any mode, it really is incredible.

There are also some more professional plates like the BladeRF that has a more sensitive receiver, and besides receiving, it can transmit. With that receiver, we can mount our station and tour the bands, we can find many things varied! Different sounds, alarms, beacons, local police, firemen, etc.

Reception of a military radar in 260Mhz, from my house.

It is so much that these receivers became popular, that many subreddits have appeared in this respect like RTLSDR to talk about the experiments that one realizes with SDR, and Signal Identification that is a specific subreddit to identify signals that we find in the spectrum.
How to receive low-orbit satellites

With this new technology, the reception of satellites lowered its costs to something insignificant, like the 8 dollars of an SDR receiver and perhaps, 200 Argentine pesos to make the antenna.

I will try to break down what is necessary to be able to build our reception station.


Orbitron: for the prediction of satellites. We can also use a website that does the same, but online.
SDRSharp: the reception software, with which we take advantage of our SDR dongle.
WXToIMG: the software for decoding satellite weather signals.


Any SDR receiver, USB dongle type
A good antenna for these cases.

And … Which antenna is good for these cases?

My favorite antenna by far, is the famous circular-polarized helix quadrangular antenna. Sounds ugly eh?

There are many ways to build a helical quadratic antenna (O QFH antenna). My favorite is the version that is made with copper pipes. We can find a manufacturing tutorial is this link.

My QFH antenna was not very neat, since it did not have a tool to bend the copper pipes, so I made the curves with a clamp and a support. Surely later on make a new, more detailed version, but it works perfectly.

The approximate cost was 220 pesos, between 5 meters of copper pipe and a tube of 2 “of PVC, all bought in the Easy.

I used the cable from my dongle to make the receive line from the antenna to the receiver. You can see the antenna installed on the terrace of the building

After some tests I started working at the reception, after work, I went home and predicted some orbit that would be useful. When living in Federal Capital, the radio stations, their harmonics and other electrical noise complicate much the work of receiving satellites, because the signal of the satellite must be above the noises with which I have to deal.

Reception of NOAA-18

First reception with an SDR dongle and an incorrect antenna, you can see the south of Argentina and the Falkland Islands.

Living in the countryside, or on the coast, the images are much better quality, cleaner and longer. Here in the big city, the buildings also play very against in passes that do not have too much elevation. The higher elevation, the better reception we will have.

How do NOAA satellites sensors work?

NOAA-15 in a 51-degree elevation run

The main imaging system of these satellites is based on an AVHRR radiometer. The radiometer is a scanning scanner, broadband, which measures the radiation in different areas of the electromagnetic spectrum, by obtaining several images of the same area at different wavelength, you can do a multispectral analysis to define parameters of meteorology, Soils, oceans, etc. By the nature of the sensor, it does not matter whether the satellite is overshadowed or not (whether it is in sunlight or in the dark). On the NOAA-15, 18 and 19 satellites the sensor is an AVHRR / 3 with 5 main channels.

Channel 1 – Visible
Channel 2 – Near Infrared
Channel 3a – Near Infrared
Channel 3b – Medium Infrared
Channel 4 – Far Infrared / Thermal
Channel 5 – Far Infrared / Thermal

The problem of elevation

As the satellite travels over the surface of the earth, from a fixed point the elevation can be calculated and its azimuth (in degrees), which is no more than the angle to the earth, and the angle to the north.

It is very important to take these data into account when performing the reception of a satellite, because if the elevation is not enough, we will have problems to listen to it. Naturally, the received images start and end in noise, which is the moment in which satellite has the least receptive elevation. Of living in a capital city, the elevation that the satellite takes is of vital importance, taking into account always, from where it is tried to receive. So also do not neglect the azimuth of the satellite (the degrees, clockwise from the North), since if we have a pass that its azimuth coincides with a heap of buildings higher than our point of view, we will have A reception of very poor quality until the satellite has enough elevation to overcome these obstacles, making a reception perhaps too short.

NOAA-15 in a pass of very low elevation, something of 20 degrees, much information was lost due to the low signal received, but still it perfectly looks the contour of the earth, and clouds.

NOAA-18 in a 35-degree elevation run

Final Words (for now)

As you have seen, the reception of satellite signals has been substantially simplified and improved. In addition to the Internet, today we have available technology at a ridiculous price that allows us to play with this technology marvel that passes over our heads.


2 thoughts on “DIY spying on Satellites

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s