Digital SSTV Decoding

I’ve decoded analogue SSTV transmissions before (check out some earlier posts) using the MMSSTV program, but another form of picture transmission is referred to as Digital SSTV.. It isn’t technically slow scan the the SSTV part has stuck because it sends images. In very basic terms it is file transfer using DRM “Digital Radio Mondiale” encoding.

The advantage over analogue SSTV is the use of error correction, with the error correction you can get a perfect image.

This afternoon while the snow was falling outside I had got SDR# with the FCDP+ running and noticed that there was some activitiy on 14.233 Mhz and fired up the EasyPal software and decoded a few images, and made a small video showing one image being received.

As you will observe in the video Easypal actually decoded the image before the end of the transmission because it didn’t need the extra data to error correct because of the very strong clear signal.
These were the two nice pictures I decoded today from OE3AWA based in Austria

Interestingly the only other previous D-SSTV image I have received was at the end of December last year, from the same operator!

Decoded some Hellschreiber

While tuning across the HF bands tonight I stumbled upon this strange signal on 40m, firing up Multipsk I took a hunch that it was some form of teletype transmission and discovered it was in fact Hellschreiber.

You can read all about it on Wikipedia

The original mechanical Hellschreiber facilimile machine was first developed in the late 1920s, and saw use starting in the 1930s, chiefly being used for land-line press services. During WW2 it was sometimes used by the German military in conjunction with the Enigma encryption system. In the post-war era, it became increasingly common among newswire services, and was used in this role well into the 1980s. Today Hellschreiber is used as a communication mode by amateur radio operators using computers and sound cards; the resulting mode is referred to as Hellschreiber, Feld-Hell, or simply Hell.

Sorry about the poor quality of the video, hopefully you can make out some of the text in this QSO.

I think I might be mad!

I know it is difficult to believe but Spring is officially here! At 11:02 today the vernal equinox occurred.

Well earlier at 07:00 I was standing out in a snow shower with the 2m yagi monitoring the ISS downlink on 145.800MHz. This was a school contact with Australia being operated via a telebridge using an Italian ground station. Details here

I did get a reasonable signal during the pass but sadly it was marred yet again by interference.

The weather is dreadful at the moment, the picture above shows the view out of the shack window last Sunday afternoon. I had hoped to capture various satellite transmissions but the weather was atrocious. It rained most of the day and then around 14:00 it turned into a short lived heavy snow shower.

On Monday night I went to catch the evening pass of STRaND-1 and AAUSAT3 only for it to start raining once I had got set up!

WSPR using a FUNCube Dongle PRO+

For sometime I have been using my FUNCube Dongle Pro+ SDR as an HF receiver station for WSPR, so. I thought it was about time I posted something.

By its very nature radio propagation isn’t totally predictable so someone transmitting can never know exactly where their signal will be received. There is whole science behind radio propagation prediction and amateur radio operators are always on the look out for openings or skip conditions for DX communications. To aid operators a number of propagation beacons exist, usually operating in CW mode transmitting their identification (call sign and location). Some of them use frequency shift keying and some transmit signals in digital modulation modes.

While invaluable operators have to actively receive and monitor these signals and what they really want to know is how their signal is getting out to the rest of the world. This is where the WSPR system comes into its own. The WSPR system uses a protocol which probes these potential propagation paths using low-power QRP transmissions.

WSPR (pronounced “whisper”) stands for “Weak Signal Propagation Reporter” and is a computer program that enables amateur radio stations to participate in a world-wide network of low power propagation beacons. The station transmits beacon signals and receives signals from other stations operating in the same amateur band. These stations then upload ‘spots’ that they receive in real time to a central website enabling operators to find out where and how strongly they were received, and can view the propagation paths on a map.

It is also possible to operate a receive only station uploading ‘spots’ to the same website, all that is required is a receiver capable of receiving single side band transmissions and feeding the resultant audio into the WSPR program where it is processed. The WSPR program was written by Joe Taylor, K1JT.

These “whisper” signals are often barely audible but their presence can be detected by the WSPR program using signal processing. The WSPR signal uses frequency shift keying (FSK) with a very small shift and a very slow data rate. The signals bandwidth occupied is only about 6 Hz so many stations can operate within the 200Hz WSPR window without interference. WSPR transmissions are encoded to carry a station’s callsign, grid locator, and transmitter power in dBm. The program can decode signals with S/N as low as -28 dB in a 2500 Hz bandwidth.

Each transmission lasts for just under two minutes, and starts at the beginning of each even-numbered minute. Therefore it is vitally important that transmitters and receivers are synchronised, so one of the fundamental pre-requisites of success with WSPR is an accurately-set computer clock. This is achieved by using internet or GPS time synchronisation methods.

Setting Up

This diagram shows the set up I am using at the moment. I have a long-wire antenna connected to the FUNCube Dongle Pro+. I am using SDRSharp (SDR#) to operate the FCDP+ and the resulting audio output is then used as the input into the WSPR program.

I am using SDR# but any suitable SDR program could be used, I have used SDR-Radio and HDSDR but I have found the SDR# program uses less resources on my ageing PC.

Routing the sound output from one program to be the input into another can be problematic and depends on the soundcard and its driver, you might be lucky and have a ‘stereo-mix’ or ‘what-u-hear’ option to use the main sound card output as a recording input, or alternatively you will need to use something like virtual audio cable VAC.

Since the WSPR signal is very narrow band it is desirable that your receiver is accurately calibrated. Most SDR program that support the FUNCube Dongle PRO+ allow a correction setting so that the tuned signal is at the correct frequency, the use of beacons, repeaters, time signals or broadcast stations is an excellent method to set this correction if required.

You will need to download the WSPR program from the current stable version is WSPR-2.11. The installation is straightforward and when you start the program a window will appear that looks like a command prompt, don’t close this window, it will give important debug messages if there problems. Then the main console window will open along with a window where you set the station parameters.

Don’t worry about the Callsign and Grid locator at the moment, the first thing to do is select the correct audio in source, the rest can be ignored as it relates to a transmitter and I am describing how to set up a receive only station.

Firstly slide the Tx fraction (%) to zero, since you won’t be transmitting and make sure the upload spots is unchecked at the moment, then select the appropriate band you wish to receive. Then set the appropriate dial frequencies in your SDR receiver program, this is shown in the window and your SDR program will need to operate in USB, with a bandwidth of 2500Hz.

The current WSPR dial frequencies are (MHz)
0.136, 0.4742, 1.8366, 3.5926, 5.2872, 7.0386, 10.1387, 14.0956, 18.1046, 21.0946, 24.9246, 28.1246, 50.293, 70.091,144.489

I suggest you turn off any AGC and any filtering in the receiver, uncheck the idle box in WSPR and then wait for the next ‘even’ minute at which point the program should show receiving, alter the sound level so the Rx Noise is ideally around 0dB, but it will work between -10dB and +10dB.

When the two minute interval is over a segment will appear in the waterfall in the top panel and the program will decode any WSPR transmissions received, you will see them as lines in the waterfall as the above image shows. Any successful decodes will appear in the bottom panel.

Once you have got it working, the next thing is to register on the website for a SWL callsign, mine is G-SWL10 you will also need to know your grid-locator, you can find this easily using

Once registered then putting the data into the station parameters and checking the upload spots will send your spots to the website database, where you can view your spots in the database and on a map, the map at the top shows my spots on one day this week on 20m, getting stations from Australia, the Far East and the US as well as Europe.

Some important things to note are ensure you computer clock is accurately set, if it is wrong you will be out of sync and decodes won’t happen and ensure you are tuned to the correct frequency as you have set the WSPR program otherwise spots will be reported for the wrong band.

It is a fascinating activity and even as just a receiver you are offering a valuable service to amateur operators.

Receiving LES1 a satellite built in 1965!

A few weeks ago I read a news article about the miraculous reactivation of LES1, this satellite was built by the Massachusetts Institute of Technology and launched back in 1965. The satellite failed to reach its intended orbit owing to a wiring error and has been drifting out of control ever since, it was abandoned in 1967 as a piece of space junk but has seemingly begun transmitting again after 46 years.

Back in December Phil Williams G3YPQ from Cornwall noticed a peculiar signal and after some research determined it was LES1. The odd signal is probably caused by the satellite tumbling end over end every 4 seconds as the solar panels become shadowed by the engine. ‘This gives the signal a particularly ghostly sound as the voltage from the solar panels fluctuates’ Phil says. It is likely that the on board batteries have now disintegrated and some other component failure has caused the transmitter on 237Mhz, to start up when its in sunlight.

Eager to see if I could receive this ancient satellite I dug out an unused DAB 3-element antenna and mounted it on a tripod and successfully monitored two passes this afternoon.

I used my FUNCube Dongle (original version) and SDR-Radio V1.5 and this was the analysis of the resulting recorded IQ file for the pass that started around 17:30pm. You can clearly see the satellite signal as it undergoes Doppler shift and the four second regular frequency shift caused by the tumbling.

At its most northerly latitude the satellite orbit passes over North Africa, but due to its height has a longer orbit period than most other satellites and is receivable over much of Europe. How long it keeps transmitting is a mystery.

STRaND-1 – I think I have a faulty antenna!

Over the weekend I have made a couple of attempts at receiving the STRaND-1 satellite telemetry and indeed have a few IQ wave files that require processing but I don’t expect to get any successful decodes out of them as they have been rather patchy and variable.

During the pass on Sunday evening I wasn’t getting anything and as it neared the end of the predicted pass I gave up and picked up the antenna tripod to put it away and suddenly a strong signal appeared on the SDR waterfall! I suspect I have a problem with the antenna, either a faulty connection in the coax or connectors that was briefly corrected by the act of moving the tripod. It isn’t the first time I have had issues and will have to investigate further.

The weekend wasn’t a total loss, I did manage to capture a reasonable NOAA-18 pass on Sunday afternoon! Between washing the cars and doing some gardening!

STRaND-1 – First attempts

As I posted earlier this week a number of satellites were launched into orbit on board the PSLV-C20 rocket, this included the first UK CubeSat, STRaND-1.

STRaND stands for Surrey Training, Research and Nanosatellite Demonstration and STRaND-1 is hopefully the first of a long line of STRaND nanosatellites from the academic researchers at the Surrey Space Centre (SSC) based at the University of Surrey in conjuction with Surrey Satellite Technologies Limited (SSTL).

The innovative STRaND-1 CubeSat was built and tested in just three months and is designed to demonstrate the feasibility of using cheap smartphone electronics to control a spacecraft, since it contains a Google Nexus One Android phone as part of its experimental payload. STRaND-1 carries an amateur radio AX.25 packet radio downlink on 437.568 MHz using 9k6 bps FSK modulated data HDLC frame, NRZI encoding.

Official orbital elements (TLEs) have already been produced so the current position and prediction passes can be calculated and available on

At the time of this post STRaND-1 is passing over in UK in the early morning (South-North) and early evening (North-South), both times are not particularly convenient on weekdays due to work commitments, but last night I did manage to have my first proper attempt to capture some of the telemetry.

My ‘ground station’ consisted of a FUNCube Dongle and SDR-Radio (V1.5) with doppler correction enabled. It was connected to a small Moonraker ZL-Special 7-element antenna mounted on a small tripod with my Android phone running the Satellite-AR application mounted behind it. The Satellite-AR assists in pointing it in the correct direction as the pictures below demonstrate.

This slightly blury picture shows the satellite cluster containing STRaND-1 displayed in the Satellite-AR app, by moving the antenna on the tripod I attempt to keep this in the centre of the screen hopefully maximising the signal.

As you can see on the waterfall I did manage to receive the telemetry signal. It isn’t a continuous signal, data is broadcast in short bursts with long gaps between the ‘frames’

The next stage was to take this and using a number of software packages, including a TNC modem emulator and extract the data. The process is described here by Jan van Gils (PE0SAT) I have had a few attempts with my captured signals, unfortunately with little success I think due to the signal to noise ratio being too low.

This morning I was up nice and early to capture the pass at approximately 05:50, this time I used my FUNCube Dongle Pro Plus on another laptop, but for some reason the received signal was very poor. In desperation I took the antenna off the tripod and held it horizontally rather than vertically and did manage to get a signal, unfortunately it was at the end of the pass and STRaND-1 was heading over the horizon.

Still I am happy with my initial efforts.