Einführung SDR: Kompakt, praxisnah, verständlich

Wer eine konzentrierte, praxisnahe und verständliche Einführung in die Technik Software-definierter Empfänger (SDR) sucht, der findet alles dazu in einem 28-seitigen und deutschsprachigen PDF von Hayati Ayguen.

Nach der spannenden Lektüre kennt man die Chancen ebenso wie die Grenzen von SDRs, kann die Prospektdaten und vollmundigen Werbeversprechen vor allem der großen Hersteller von Amateurfunkgeräten besser einordnen und lernt somit auch die Leistung sowie den Funktionsumfang der Produkte kleinerer Hersteller noch stärker schätzen.

Einen ersten Überblick bietet Hayati auch auf Folien.

INMARSAT: Decoding 12 Aero-channels in parallel

Jaero12

Action: Free software allows for decoding twelve INMARSAT in parallel

A recent post in Carl’s rtl-sdr-blog informed about the ebay-lability of some surplus Outernet patch antennas for just – see here. For just 29 US-$, I got this small antenna with integrated SAW filter (1525 – 1559 MHz) plus LNA. A real bait for me to jump over the limit of 30 MHz reception! Soon I fired up my AirSpy R2 receiver, providing the LNA with power supply (Bias-Tee). It worked fine, and I received a whole bunch of excellent signals by this setup.

As I wanted to receive some aircraft information, so I downloaded free JAERO decoder of Jonathan “Jonti” Olds, also from New Zealand. This fine software can be opened in many instances. In combination with the up to 24 decoders of SDR-Console V3 of Simon Brown, this modest setup turned into a multi-channel satellite reception post.

AeroGUI

Here 12 decoders had been assigned – one on each INMARSAT channel. You see also quite good SNRs from the Outernet patch antenna.

Next steps worked as usual with the mutli-channel approach:

  • make up 12 channels in SDR-Console and tune each channel to a different signal. Mode must be USB, and as bandwidth I choose 1200 Hz for 600 bps and 2400 Hz for 1200 bps channels. That’s a bit wider than necessary, but doing so there is some room for the AFC in JAERO decoder always to stick to the signal even if the SDR should drift a bit over 24 h or so
  • The output of each channel is then routed to a different Virtual Audio Cable, or VAC 1-12.
  • Then you have to install twelve instances of JAERO software in different folders, e.g. JAERO 1-12. You should name each JAERO.exe file accordingly, e.g. JAERO_1.exe to JAERO12.exe.
  • Open JAERO_1.exe, assign its input to VAC 1, and set the matching speed of the signal. If all is ok, you will be rewarded by a sharp phase constellation, and soon decoding will start.
  • Repeat the above steps until you have reached JAERO_12.exe, connected to VAC 12.
12Matrix

The “Matrix” of SDR-Console V3 shows the twelve channels with different signal strengths and width, depending on the data rate (600bps/narrow, 1200bps/wide).

The result can bee seen from the screenshot at the top of this page. The whole setup ran stable and unattended for hours.

Thanks for all smart people having developed the smart software and hardware!

Fighting for Annobon, 3C0W

3C0W_10119_annotated

3C0W is working CW, split operation, on March13th, 2018. After showing up on 10.119 kHz, a pile up starts.

Ham radio operators under a rare call sign mostly work “split frequency” to efficiently serve as many stations as possible. The screenshot on the top shows this in practice: 3C0W from Annobon starts calling “up” on 10.119 kHz on the bottom of this spectrogram. Within minutes, a pile-up builds up, getting stronger in the number of stations as well as in width (“spread”). On the top of the picture, 3C0W requests a break (QRX), and most calling ceases.

The picture at the bottom shows just a 20 seconds’ clipping of this session. It is an excellent illustration of a sentence by the late L.A. Moxon, G6XN (author of e.g. “HF Antenna for all Locations”, RSGB Potters Bar, 1982, p. 216): “Communication in the amateur bands is usually restricted by interference rather than noise so one might argue that little would be lost and much gaines from a power limit of 30W, universally applied”. And even back in 1924, S. Kruse, the then-Technical Editor of ARRL’s “QST” wrote (May issue, page 36): “Practically none of us paid the slightest attention to making the signal redable or steady; no, the main idea was to make a noise.”

As you see, many problems still remain after nearly a century: people are calling when the DX station is transmitting and definitely cannot listen, and some station overdrive their transmitter in a way that they nearly bury the DX signal.

3C0W_10119_detail_annotated

Annotated detail of this pile-up: It shows overdriven signals calling even when 3C0W transmits and cannot listen …

SDR-Netz des DARC e.V.: Rauschende Ergebnisse

R2T2_Braunschweig_ANT1

Einfach vergleichen: Oben CHU 14.670 kHz an einem Remote-Standort des DARC e.V., unten zur selben Zeit an einem durchschnittlichen Standort.

Aus den Mitteln seiner “Mitgliedschaft Pro” bestellte der DARC e.V. im Jahre 2014 sein Web-SDR-Transceivernetz. Es soll seinen Mitgliedern “weltweiten Funkbetrieb aus dem heimischen Shack ermöglichen”. Jeder der über 1.000 DARC-Ortsvereine war aufgerufen, sich als einer der zwölf Standorte zu bewerben. Gesucht waren solche Locations, die vor allem einen störungsarmen Empfang und gute Antennenmöglichkeiten bieten – was der Funkamateur in der Stadt eben nicht hat.

Obwohl die R2T2 genannten Geräte für über 25.000 Euro längst ausgeliefert wurden, ist es bedauerlicherweise merkwürdig still um dieses schöne Projekt geworden; auch die zugehörige Yahoo-Newsgroup scheint nicht mehr ansprechbar zu sein.

Wie also ist der von außen (ich bin kein Mitglied des DARC e.V., begrüße aber dieses Vorhaben uneingeschränkt!) sichtbare Stand des Projektes?

Um das zu erkunden, habe ich am 20.12.2017 alle verfügbaren Remote-Standorte (sechs, und die auch lediglich empfangsseitg ansprechbar) mit dem Empfang “im heimischen Shack” – leider einer ziemlich durchschnittlichen Location – verglichen.

Die ersten Ergebnisse habe ich in einem PDF zusammengefasst, das ihr hier unter R2T2 herunterladen könnt.

Diese Versuche wurden zur Vervollständigung des Bildes fortgesetzt – siehe unten. So nahm ich am  21.12. um 07:45 UTC einen Vergleich auf 17.950 kHz vor, wo bei mir der Rundfunksender China Radio International/Kashgar mit einem SNR von gut 31 dB einfiel: bis auf ein sehr schwaches und praktisch unverständliches Signal vom Remote-Standort Wiblishauserhof war auf den anderen Remote-SDRs des DARC so gut wie nichts zu hören, zum Teil wegen (lokaler?) Störungen. Unten der Vergleich meiner Station (unten)  mit dem bayerischen Remote-SDR (oben).

r2t2_Bayern_17950

Vergleich China Radio International, 17.950 kHz: oben der bayerische SDR des DARC-Remote-Netzes, unten dieselbe Station zur selben Zeit am Standort DK8OK.

Einen weiteren Test unternahm ich am 26.12.2017 gegen 12:30 UTC auf 1.521 kHz (CRI/Kashgar, Nähe zum 160-m-Band) und auf 4.800 kHz (China National Radio 1/Golmud). Diesmal erfolgte der Vergleich an einer Aktivantenne (statt der Quadloop)  meinerseits, die bei beiden Stationen eine Empfangsqualität von SIO 253 bot.
In beiden Fällen war das Ergebnis ähnlich: von den sechs verfügbaren Remote-SDRs konnte drei die Stationen überhaupt nicht empfangen. Eine weitere lag knapp über der Hörschwelle, Schöppingen zog fast gleichauf, während Wiblishauserhof in etwa Gleichstand mit meiner Anlage bot – als einzige Station von zwölf bezahlten/geplanten. Für gut drei Jahre Bauzeit und über 25.000 Euro Investment ein Befund mit durchaus Luft nach oben.

Eine Fortsetzung derartiger Vergleiche scheint daher solange sinnlos, wie das DARC-Netz nicht erweitert bzw. Standort/Antennen/SDRs entscheidend geändert werden.

Was hingegen engagierte Hobbyhörer ehrenamtlich und mit allein privatem Geld im Gegensatz zum “Bundesverband für den Amateurfunkdienst” der staatlich geprüften Hobbyfunker zustande bringen, zeigt – ebenfalls mit 14-Bit-SDRs – das leistungsstarke Kiwi-Netz.

Airspy HF+: What you hear, is what you get

 

 

This slideshow requires JavaScript.

It has been dubbed “game changer” and indeed, the Airspy HF+ is a completely other animal of software-defined radio, or SDR. Developed by Youssef Touil plus team and produced by ITEAD, it sells for just US-$ 199 right from factory at Shenzhen, China. This is considered the middle class of SDRs, starting with cheap USB sticks under 10 US-$ and scratching the mark of nearly US-$ 5.000 with Winradio’s WR-G39DDC. This one also marks the transition zone from what even an engaged hobbyist allows himself to spend to the truly professional receivers of e.g. Rohde & Schwarz and Plath. To make it clear: You may achieve professional results at each price tag, even from an RTL & its clones, as Carl Laufer’s excellent blog shows almost daily.

The Serious HF-DXers in mind

Airspy HF+ has been developed with the serious HF (shortwave) listener in mind. In this field it sets new standards regarding sensitivity, dynamic range and noise. Its stunning performance is achieved by a revolutionary approach and a careful layout of the hardware, housed in a sturdy metal case.

I don’t want to add another explanation of this concept (my test report will appear in 1Q/2018 in “Funkamateur“) but just offering the pure stuff. Some first twelve audio examples should give you a truly hands-on impression to answer the one and only question: How loud does this animal roar?
Therefore, I compared about 100 often vastly different situations on HF between Elad’s FDM-S2 (US-$ 525) and Airspy HF+. From this collection, I carefully selected some first twelve examples to cover the needs of the casual listener as well as the hard-core DXer.

All audio clips were recorded in parallel with a 20 m quad loop as antenna, feeding a professional 1:2 HF splitter by Heros. Software used was free SDR-Console V3 by Simon Brown – thanks.

Dare to make use of your own understanding

First, you read a description of the case, followed by a recording with FSM-S2 and then by Airspy HF+. Each of both examples has been recorded with exactly the same bandwidth, mode, AGC etc. which had been optimized for that situation. You must listen to these audio clips with headphones to scrutinize the mostly very small differences. Aim you ear towards fading, noise and intelligibility.
This is not a traditional test, where the master of ceremonies masticates the results for you. It’s for the truly demanding DXer, “to make use of your own understanding” (Kant, 1784). Just a hint: weak stations make the difference!

Fasten your Headphones: The Examples

The audio examples are roughly sorted from easy to difficult signals. They were made in the first week of December, 2017.

Radio Sultanate of Oman, Seeb/Oman
15.140 kHz, 100 kW, AM, 5.350 km, 14:10 UTC, strong/free channel, SAM, 10 kHz bandwidth. Keep an ear on noise and slight fading!

 

Xinjiang People Broadcasting Station, ÜrümqiChina
4.500 kHz, 50 kW, AM, 5.500 km, 14:24 UTC, fair to good/free channel, SAM, 9 kHz bandwidth.

 

Bangla Desh Betar, Savar/Bangladesh
4.750 kHz, 100 kW, AM, 7.300 km, 14:29 UTC, fair/free channel, SAM, 9 kHz bandwidth.

 

Xizang People’s Broadcasting Station, “Holy Tibet”, Lhasa/China
6.025 kHz, 100 kW, AM, 6.850 km, 16:00 UTC, fair/strong broadcaster 5 kHz up, ECSS-L, 2,8 kHz.

 

Bangkok VOLMET, Bangkok/Thailand
6.676 kHz, 10 kW, USB, 8.800 km, 16:10 UTC, fair/free channel, USB, 3 kHz bandwidth.

 

Gander VOLMET, Gander/Newfoundland Canada
10.051 kHz, 10 kW [?], USB, 4.400 km, 15:20 UTC, weak to fair at fade-in/free channel, USB, 2,8 kHz bandwidth.

 

Myanma Radio, Yangoon/Myanmar
5.985 kHz, 50 kW, AM, 8.250 km, 01:00 UTC, weak to fair/interference from upper channel, ECSS-L, 5,5 kHz bandwidth.

 

Radio Aparecida, Aparecida/Brazil
6.135 kHz, 10 kW, AM, 9.900 km, 00:30 UTC, fair/free channel, SAM, 3,5 kHz bandwidth.

 

Time Signal Station CHU, Barrhaven/Ontario Canada
3.330 kHz, 3 kW, USB with carrier, 5.900 km, 06:00 UTC, fair/fsome interference from digital station above, USB, 3 kHz bandwidth.

 

Time Signal Station BPM, Shaanxi/China
15.000 kHz, 20 kW, AM, 7.750 km, 09:00 UTC, weak/free channel, SAM, 5 kHz bandwidth. Occasionally echo from mixing short/long path, some CW echo (long path) is running into the next via short path.

 

China Radio International, Ürümqi/China
1.521 kHz, 500 kW, AM, 5.500 km, 13:00 UTC, weak at fade-in/free channel, SAM, 6 kHz bandwidth.

 

Auckland VOLMET, Auckland/New Zealand
6.679 kHz, 5 kW, USB, 25.800 km (long path!), 07:20 UTC, very weak/free channel, USB, 3,6 kHz bandwidth. Here headphones are a must!

 

 

Offset/SNR: Some Ideas for Medium Wave DXing

Offset_5

Offset-DXing “on the fly” shows four different stations (spectrogram) on one nominal channel, namely 801 kHz. The window is baout 30 Hz wide and shows the carrier on HF level.

Although I use Simon Brown’s excellent software SDR Console V3 for years, I only now discovered a feature, being most valuable for medium wave DX.

Nearly each medium wave channel is populated by a couple of stations which mostly have a slight difference from each other, called offset. This often is specific to specific stations. It even reveals stations too weak to be heard. Software V3 will show these carriers of HF level during normal listening, being live or from an HF recording.

Read MW-Notes, to get some information on “how-to” on 6 pages, with 12 screenshots. There you will find also a hint for a method with even much more resolution (but: not “on the fly”) plus some information on how to measure signal strength and estimate/calculate the SNR of speech/music, rather than that of just the carrier.

You have to distinguish between absolute and relative frequency accuracy; the first is best achieved with a GPS-disciplined oscillator, the letter the normal case.

P.S. I started with these things back in 1997 with an evaluation board from Motorola, followed by sound card & software on audio level (“Soundtechnology zeigt Signale: Sieh’, wie es klingt!”, funk magazine 6/1998), to be continued on HF level from 2006, first with RFSpace’s groundbreaking SDR-14. Three years later, I published a survey of each and every 9- and 10-kHz-channel on medium wave by this method. After Apple closed their web service, these pages had gone astray, and the information is now not up-to-date anymore. State-of-the-Art now is the method described in the paper.

Medium Wave: Ex oriente DX

Medium wave saison has started, and am I trying to make the best out of it. Conditions are fascinating different from day to day, and even from minute to minute. With mainly focusing on “East of Suez”, with some other in between, please find some 50+ audio logs below. Click “Read More” button at the end, to expand the list to full length.
I am very much indebted to Christoph, OE2CRM, who with his very special mixture of charme and nuisance more or less forced me to explore a bit more of this frequency range 😉 First of all, I was and still am attracted by his outstanding logs which had been held impossible in Mid-Europe in the last decades.
I am using an Elad FDM-S2 at a wire loop of 20 m circumference with Wellbrook’s Large Aperture Loop Amplifier ALA100LN plus 7th order elliptic low-pass filter (1,5 MHz) by Heros to avoid any spilling over from HF (mainly that of: Radio Romania International); software used V3 from Simon Brown.

BEL3_Fishery_FRS_1143kHz

Part of the QSL from “Fishery Radio Station” (Taiwan Chü Yuyeh Kuangpo Tientai), BEL3, 100 kW, 1143 kHz, received September 25, 2017, 19:00 UTC. 謝謝, Station Manager Jin Mey Ju!

1700 kHz USA-Florida  WJCC Radio Mega in French, Miami Springs, 10 kW, 10-OCT-2017, 02:00 UTC.  Several IDs (e.g. in French) of this multi-cultural broadcaster.

 

1584 kHz G  Punjab Radio, in Hindi/English, London, 2 kW, 15-AUG-2017, 20:00 UTC. ID.

 

1566 kHz KOR  HLAZ FEBC in Korean, Jeju, 250 kW, 26-SEP-2017, 17:00 UTC

 

1566 kHz HOL  Vahon Hindustani Radio in Hindi, Den Haag, 1 kW, 23-AUG-2017, 22:00 UTC. ID in Hindi.

 

1557 kHz TWN  RTI iLoveMusic in Chinese, Kouhu, 300 kW, 20-SEP-2017, 16:55 UTC.

 

1550 kHz ALG  Saharawi Arab Democratic Republic National Radio in Arabic, Tindouf, 50 kW, 27-SEP-2017, 21:01 UTC. ID: “RASD punto [?] info …” & in Arabic

Read more

« Older Entries