Frustrating...

[DaveNF2G]

I suppose I am turning into one of those old codgers.

I find the detection and decoding of digital modes on HF to be extremely frustrating of late. I used to be able to attach a simple decoder box to my receiver and decode RTTY at 3 different speeds and various shifts, ARQ/FEC (AMTOR/SITOR/etc.), and tell them apart by ear. Another box took care of WEFAX and my PK-232 even handled NAVTEX.

Now it seems there are a gazillion modes, many of which sound alike. When I find a digital signal, I have an arsenal of decoding programs that generally cannot even figure out what sort of signal it is, let alone decode it. DRM decoding is still a DReaM for me, despite having that software and occasionally being able to find DRM transmissions on published frequencies at scheduled times.

Last night I found a signal that, according to all the usual resources, was clearly STANAG. I ran the Sorcerer program against it and could not even confirm that it was STANAG, or which variant it might have been. No luck at all determining the status of the transmission (idling or carrying traffic) or confirmation that it was encrypted.

(BTW, to my ear, STANAG and DRM sound almost exactly the same, so I try both programs against them before declaring failure.)

I have found several signals that sound exactly like Baudot RTTY, but they steadfastly resist decoding as well. Only the amateur modes work somewhat reliably, along with FAX. Everything else is just noise.

BTW, I get the same results whether I run a wire from an actual receiver to the jack on the computer or use an SDR.

 

[ka3jjz]

No you're not one of the old codgers - just not quite up on some of the many digital modes that are not amateur related. There are quite a few of these, and some are encrypted. STANAG signals are often encrypted.

However, to at least address some of the frustration, try tuning these stations (which have regular skeds and/or frequencies) - AFAIK it's not known yet how the USCG consolidation (see the message in the utility forum) will affect the SITOR-B broadcasts.

Testing Your New Setup - The RadioReference Wiki

And as an aside - you might also find several SITOR-B stations that broadcast in languages other than English...and even users like the Egyptian embassies have been known to use SITOR-A and B...

Now trying to tune an unknown dig signal is a little more work. Some sites - like the SigIDwiki that was mentioned in the UDXF a bit ago - have audio traces (some call them waterfalls or spectrograms) of many different modes. This is one way to go, but it's far from an exact science; the accuracy of the waterfall will be affected by how clean the signal is, as well as any interference (man made or natural). There are a great many software packages - some of them free - that can show waterfalls at various levels linked at the bottom of the UMC page in my sig.

I know of quite a few folks on the UDXF that use fairly detailed analysis tools to determine whether a signal is encrypted - and some have very unique characteristics that help define the mode. This article describes some of this in detail, as well as mentioning some lower-cost packages that have at least some of these capabilities..

Decoders - The RadioReference Wiki

In short you need to be something of a detective - being aware of the logs in the UDXF and Spectrum Monitor can help to a certain extent, but when you're in this field, analysis is a real key here....

 

[SCPD]

BTW, I get the same results whether I run a wire from an actual receiver to the jack on the computer or use an SDR.

Dave, I actually think you'll get better results with a SDR. You didn't specify *which* SDR you are referring to... if you mean one of the cheapy RTL SDR with a converter then that's not a quality SDR.

I suspect what you're running into is frequency stability on both your radio or cheapy SDR. Sorcerer is very picky about frequency. If you think about it, you're asking the software to crack a safe... the software can see the safe but it can't quite reach it -- so how is it suppose to do its job?

A quality SDR will make a huge difference when it comes to decoding.

For amateur modes, you should be using FLDIGI to decode. It's quite a bit better at decoding than any other program right now. The fact that it's free is a bonus!

PS. RFI of any kind can drastically affect digital decoding no matter what platform you try it on...

 

[ka3jjz]

One thing I have read on SDRs is that it's wise to calibrate them against a known source, such as WWV....

As good as FLDigi is, it won't copy a lot of the non-amateur modes that are out there today. ALE (yes, that's both in the amateur and non-amateur world) and HFDL (both pretty commonly found these days) are but 2 examples.

MultiPSK is better, but you have to get past all those buttons - Patrick evidently doesn't believe in pulldowns. That panel is really too busy and can turn folks off....Mike

 

[DaveNF2G]

I have noticed that MultiPSK's multisignal PSK31 display fills up with garbage pretty fast. Other PSK31 programs I've tried only showed something when there was a valid signal present.

 

[SCPD]

Well if you have the money to spend there are plenty of commercial software packages out there... but be prepared to put up some serious ca$h. The fact that 85% of the digital traffic is encrypted might be a turn-off.

 

[SCPD]

One thing I have read on SDRs is that it's wise to calibrate them against a known source, such as WWV....

Yes, I do this every time I use my Perseus. It takes all of 5 seconds to calibrate it against WWV.

 

[ka3jjz]

I have noticed that MultiPSK's multisignal PSK31 display fills up with garbage pretty fast. Other PSK31 programs I've tried only showed something when there was a valid signal present.

I should have mentioned this earlier - there is a MultiPSK Yahoo support group, which appears to be a pretty busy place...

https://groups.yahoo.com/neo/groups/multipsk/info

Mike

 

[ka3jjz]

If I were to build a digital software library, I'd probably have Sorcerer, Sigmira, MultiPSK, PC-ALE and PC-HFDL (in no particular order). With the exception of MultiPSK (which has a somewhat hefty registration fee), all the others are free. MultiPSK has a fairly hefty registration fee, but given the modes it opens up, if they're of interest to you, it's probably worth it.

YMMV, of course Mike

 

[ka3jjz]

Well if you have the money to spend there are plenty of commercial software packages out there... but be prepared to put up some serious ca$h. The fact that 85% of the digital traffic is encrypted might be a turn-off.

A lot of the digital logs in the UDXF have to do with ALE signals that are not link-protected. There's been a HUGE growth in these signals over the last few years. It is true, however, that there's a bunch of encryption out there - as examples, most STANAG signals are, as well as some of these PACTOR variants run by stations like Bern Radio.

Some folks like the challenge of identifying some of these Russian (and other) digitals that come out as seemingly random numbers. This is the bread and butter of the Numbers and Oddities sub-group that exists within the UDXF. It greatly depends on what it is you are trying to hear.

One thing that ST editor Mike Chace-Ortiz and I have noticed is that with many of these multi-kilobuck packages, you are getting decoding options for a fair number of modes that are no longer used, or have been used so rarely that it makes you scratch your head and wonder what's up here. I suppose you could make the argument that these no-longer used modes might show up again someday, but the huge migration to ALE driven systems seems to suggest against it.

Mike

 

[DaveNF2G]

If I were to build a digital software library, I'd probably have Sorcerer, Sigmira, MultiPSK, PC-ALE and PC-HFDL (in no particular order). With the exception of MultiPSK (which has a somewhat hefty registration fee), all the others are free. MultiPSK has a fairly hefty registration fee, but given the modes it opens up, if they're of interest to you, it's probably worth it.

YMMV, of course Mike

I have all of those and then some. I did register MultiPSK because access to VDL2 was worth it for me.

 

[Token]

One thing I have read on SDRs is that it's wise to calibrate them against a known source, such as WWV....

I find I seldom (as in 2 or 3 times a year max) have to calibrate my WinRadio G31 or G33, both of those are typically within 1 or 2 Hz of on freq. Then again they are in a temperature stable environment and are on 24 hours a day, so they do not vary much in temperature.

I never have to calibrate my RFSpace NetSDR or my WinRadio G35DDCi/XP as both of those are disciplined to a 10 MHz GPS based source. One of my 2 SDR-IQ's is also modified to be referenced to a 10 MHz GPS source.

All of my receivers I use, SDR or not, get checked for calibration several times a year. A few receivers get checked more often. RTL SDRs I check every time I use them for more than casual scanning. I calibrate with local sources, not over-air.

Also, when using over the air signals to calibrate, such as WWV, keep in mind that sky wave signals can have Doppler shift on them. WWV at 15 MHz can have as much as +/- 5 Hz, although +/- 2 Hz is more typical, there is a good study on this for reference "Doppler Shift and Spread Study for Ionospherically Propagated Signals".

So if you are trying to calibrate to a nats behind on over the air sky wave signals you may be doing yourself a disservice. Then again, how many folks need frequency accuracy to inside 5 Hz?

T!