Dream and libfaad2

One of the problems with using Dream on Windows is that it doesn't come with libfaad2, the AAC decoder library. Without it, no audio is going to be played.

I followed the recommendation to compile the sources, and was successful with MS Visual Studio C++ 2010 Express. There was no problem with the custom project file. Still, it seems like a waste to install MSVC++ 2010 just to build a single DLL.

If you don't want to go through the hassle, you can use Sodira. If you don't really like the interface (and I don't), copy the included libfaad2.dll into your Dream directory, and rename it to faad_drm.dll (and it must be faad_drm.dll, not faad2_drm.dll).

I've tried it and can confirm it works, although the DLL size is a couple kilobytes larger compared to the DLL I compiled myself (229 KB vs 232 KB). Yeah, it doesn't matter.

I also tried the pre-compiled libFAAD2 2.7 DLL from Rarewares, and it crashes Dream, so don't use that one. But you could try your luck with a 2006 build of Dream, listed at the bottom of that decoders page.

A couple notes about building a DLL, just so I can reference it in case I forget:

• Yes, you should really download faad2_project_files.zip
• I used faad2-2.7.zip, not that it matters
• Unzip faad2_project_files.zip into faad2-2.7/libfaad
• Open the 2010 solution file
• Switch build target to Release
• Build and rename faad2_drm.dll to faad_drm.dll. They could have renamed the build target if they were going to make a custom solution file.

Vatican Radio DRM @ 9800 kHz, 29 Jul 2012

When VA DRM was a no-show yesterday I thought that Sackville had stopped relaying. It would have been before the end of the month, but close enough.

Last time there was some intermodulation, but this time around I reduced the overall gain of the antenna by about 10 dB. So it does seem that I was unintentionally exceeding the antenna's IP2 and IP3 specs.

If Sackville does stop relaying Vatican Radio at the end of July, this will be the last DRM transmission from Sackville that I will hear. That leaves The Disco Palace on the weekends, and REE on the weekdays. If I'm lucky and/or willing to stay up a bit later than I ought to, I could get RNZI.

I do not have a DR111 but I've got an opinion anyway

Trying to reconcile the dismal performance of the DR111 found by PCJ Radio [Vimeo][YT] with the enthusiastic reception over at drmna and friends. There is a video series of it in action, as well as signal generator tests. From the videos, audio quality is mediocre, but it looks to be capable of pulling in DRM even if the receiver is not in the target area (i.e. RNZI on the American West coast).

But if it costs $120 USD? Not worth it. It's small enough to be portable, but you're not paying for portability since it doesn't take batteries. And if it's not going to be portable, you really are better off with a cheap laptop, cheap radio with IF output, and Dream or Sodira as your decoder.

Can't afford a laptop? Then just get the cheap radio. Analog isn't going to roll over and die any time soon.

This doesn't sound like a ringing endorsement of digital broadcast radio. It's not. If the price tag of the DR111 is any indication, DRM will be expensive for a while yet. A cheap receiver should cost, at most, $20 to be considered cheap as AM. That's probably not feasible given the amount of DSP one has to throw at a DRM signal, so let's say max $50 for an inexpensive DRM standalone receiver.

I think you can get pretty close to $50 if you didn't have to pay to license the patents of the DRM Consortium. Nothing against them personally, I'm sure they're nice people, but do you know how much it costs to get a DRM license key to use in an Excalibur? Almost 50 Euros. At current exchange rates, that's about $60 USD, which is half the cost of the DR111.

I guess that assumes that the manufacturer is paying a license fee at all, as opposed to committing patent infringement, and is paying what the rest of us plebs pay. But if that assumption is wrong, then Chengdu NewStar is making sick, Apple-like margins, and not really advancing the cause of DRM for the masses.

Anyway, DRM used to be for people who can afford more than a cheap mobile phone. After the DR111, DRM continues to be for people can afford more than a cheap mobile phone.

Digital mode robustness

I learned something interesting this evening: Turning on a CRT-based television raises the noise floor by about 5 dBm across the board.

As mentioned last post, I was monitoring a pretty interesting conversation between VE3SKY and CO2DC on Olivia in the 20 m band, getting about -13 dB SNR from CO2DC as reported by MultiPSK. That seems overly pessimistic. From the Excalibur spectrum view, though, at times the signal did appear to sink below the noise floor, and was only 2 or 3 dB above, best case.

Until the TV wiped out all but the strongest PSK31 signals, decode performance was solid, with only the odd word being completely unintelligible. VE3SKY mentioned how he rarely used Olivia, and CO2DC replied that he actually preferred it.

I've only witnessed two contacts using Olivia and I see the appeal. It's more robust compared to BPSK31, with lower latency than JT65A. From an audio point of view, it's also not as depressing as JT65A.

On the other hand, it seems as if you could go a lot faster given the bandwidths in question. While kicking around the idea of a faster mode with error correction, I stumbled across a comparison of QPSK31 vs BPSK31. Seems that even the creator is kind of disappointed over the popularity of BPSK31.

I think that whatever technical concerns about QPSK31 raised in 2002 are now moot in 2012. Even if your contemporary receiver or transmitter has a bit of drift, on the receiver end there's more than enough computing power to figure things out. If you can demod DRM with a laptop, then you can demod some low quality QPSK31 signal with the same.

But we should be able to do slightly better now. Use QPSK31 or QPSK63 with a constraint length 9 convolutional code. There's at least one open source library from KA9Q that supports it.

From there it's not far of a stretch to further fortify the transmission with a block code (Reed-Solomon, or BCH), or switch the convolutional code for a turbo code or LDPC code, computing power permitting.

Over-modulated PSK31 on 20m, 28 Jul 2012

Was loitering around 14070 kHz when this really fat PSK31 signal showed up on the waterfall. The operator was KI4MRH, who had been on for at least 15 minutes prior, but I only decided to screen capture around 23:38 UTC.

On the Excalibur waterfall there are "fingers" that jut out from the centre frequency on either side. But as seen on MultiPSK most of the interference was heavily weighted on the lower sideband. I can't think of anything that would cause that asymmetry.

On a more positive note, he made contact with UT5XR.

Other novel sightings this evening (no screen shots for either):

• AD4WG had -80 dBm signal power on both PSK31 and PSK63. First time I've ever decoded PSK63, although I have seen it before and not recognized what it was.
• A rag chew over Olivia between NM4V and N1CSR. First time I've seen and decoded Olivia. One of them was in pretty deep fade, while the other came and went, but performance was something like, 80% copy.
• Another Olivia contact between VE1SKY and CO2DC.

WR-G31DDC Excalibur UTC month bug fixed

In version 1.55 of the WR-G31DDC Excalibur the date box was ahead by a month when displaying UTC time. That has been fixed in a newer version of the receiver application, but you will have to contact WinRadio support to obtain it.

DSB @ 7290 kHz, 28 Jul 2012

Around 1820 UTC. Screenshotted just for the novelty. Could only make out one callsign, which was WB9GKZ who was not the strongest signal. Whoever that was would always fade when giving their own callsign before passing the frequency back.

Link to ARRL's band plan.

RNZI DRM @ 17675 kHz, 24 Jul 2012 03:30 UTC

What's awesome is being in Canada, hearing RNZI's Brian Clark talking about DX listening from Canada. Mentioned stations included: CBC Radio Norde Quebec, CFRX, some other CBC stations on the West coast, and Radio True North.

The program then moved on to heard stations in America, Africa (Zanzibar), and Papua New Guinea. The following segment discussed recent solar activity (shortwave dropouts two weeks ago).

SNR really began to take off around 03:35 UTC, currently signal power is S9+10 (-67 dBm), subcarrier power peaks around -90 dBm.

RNZI DRM @ 17675 kHz, 23 Jul 2012

As if on cue. Frequency selective fading is pretty aggressive, with consequent lost audio frames or the entire MSC being dropped.

And just now, that interferer harmonic around 17685 kHz decided to show up. SNR is falling off a cliff, which compels me to get some sleep.

16m from Pacific, business as usual

This sorry image is my monitoring of where RNZI DRM ought to be. The times where SNR rose above 6 dB I could get the station ID to display. All other times I'd get frame sync, at best, and now it's faded into the noise floor.

The Disco Palace DRM @ 17875 kHz, 21 Jul 2012

First successful reception of TDP, beginning at 20:32 UTC, a bit after Vatican Radio DRM wrapped up.

About 4 minutes later, transmission was shut down for about 24 seconds. See red arrow.

Vatican Radio DRM @ 9800 kHz, 21 Jul 2012

Edit: It's probably transmit spectral masking as opposed to receive intermod.

It looks like the antenna is introducing some intermodulation. I've seen this also in the REE DRM broadcast on 9630 kHz, so I wonder if it's DRM specific intermod, or just intermod due to strong signal, because that's a really strong signal. An antenna with signal gain is pretty helpful.

The intermodulation is more evident here.

Around 20:28 UTC someone accidentally switched the transmitter off or something. In the gap you can see -105 dBm interference.

Memories of Vatican Radio DRM @ 9755 kHz

The broadcast at 9755 kHz was the only DRM station I stood any chance of consistently decoding during the work week, because it occurred in the evening when, sometimes, I wasn't at work. It used 16-QAM for the main service channel, which made for solid reception at 14 dB.

Honestly, I don't know why operators don't run with 16-QAM MSC more often. Something like Vatican Radio, which is predominantly speech, really doesn't need the added bitrate that 64-QAM offers. People with cheap antennas (that may or may not have been pilfered from an AM radio) would get way better results. And by way better, I mean any reception at all.

Sackville is closer, broadcasts with more power, and yet there were weekend afternoons where I'd just stare at the waterfall on 9800 kHz and see consistent 14 dB SNR due to propagation conditions, and thus no audio due to 64-QAM.

One of the last broadcasts I received before Santa Maria di Galiera discontinued shortwave broadcasts to North America was on 2 June 2012, 2300 UTC. I remember this one specifically because it had its spectrum flipped, i.e. the pilot tones were on the lower sideband instead of the upper sideband, mirrored about the carrier. This was before I thought it was a good idea to take screenshots of waterfalls, so you'll just have to take my word for it.

That wasn't all. Within the first ten minutes, the transmission stopped and restarted. I thought they were going to fix the inverted spectrum, but no dice. They used a 64-QAM MSC, and audio bit rate was higher as well (17.46 kbps EEP AAC+ Mono). Usually it's 14.06 kbps with 16-QAM MSC. The audio stream frequently dropped out, and most if not all the content was in Spanish. The broadcast technically went about 10 minutes past its scheduled time with the DRM equivalent of a dead carrier (DRM modulation, empty audio frames).

I only recall this much because I dug up an e-mail I sent to Radio Vaticana. They never replied. Probably didn't see a point, since this was an unusual one-time screw up.

RNZI DRM @ 17675 kHz, 11675 kHz 21 Jul 2012

Around 23:30 EDT (3:30 UTC) snippets of audio can be decoded, but this is the second night where conditions for 17675 kHz rapidly improve starting around midnight, and stay that way for about 45 minutes, which is about when RNZI DRM switches to 11675 kHz.

But this time around, the nearby interferer is present during the broadcast, and it caused co-channel interference. I think it's a neighbour's LCD or plasma television.

RNZI DRM on 11675 kHz was going pretty well, until BBC Arabic @ 11680 kHz from Woofferton showed up.

So I went to sleep. It was 1 am, anyway.

Oh, one last thing. If you went with the official (?) DRM broadcast schedule, you would have thought RNZI jumped to 13730 kHz beginning 04:59 UTC. This contradicts RNZI's own schedule.

Digital Radio Mondiale on WiNRADiO

The WR-G31DDC comes with a built-in Digital Radio Mondiale (DRM) decoder. Unfortunately it costs $49.95 USD Euros to unlock. Now it can be said that if you're willing to shell out for this receiver, then you can afford to pay to unlock the DRM part. Then again, it only decodes DRM. I coughed up $49.95 USD for the Virtual Sound Card instead, and pipe the VSC into a free external decoder.

There are free or at least cheaper and more general purpose virtual sound card drivers out there, but for me the integration into the receiver GUI made it about as convenient as the DRM decoder, while being more general purpose.

The Dream DRM receiver for Windows is most happy with an input signal that is 48 kHz mono, with an average power of -60 dB. -40 dB is the upper limit on its power chart, and anecdotally I've found that Dream loses sync when the signal is that strong, so it's probably clipping.

How to get that -60 dB target required some trial and error. I ended up with slow AGC , target level of -21 dB. An alternative strategy, if the signal experiences really fast fading, is to just impose a manual constant gain, and let Dream's post-FFT channel estimation do its work unassisted, but picking that gain require some trial and error.

For tuning and RF demodulation, I use upper sideband, 10 kHz of output audio bandwidth, and the upper sideband demodulator placed 5 kHz below the (suppressed) carrier.

Some comments on Dream settings:

• MLC correction = 4: Supposedly there isn't much performance improvement with increased iterations, but given my circumstances I probably need as much advantage as possible.
• Bandpass filter = ON: Given the interferer that bleeds into 9800 kHz, a BPF is a good thing, and I can live with some slight attenuation.
• Modified metrics = ON: According to this, performance is impaired when switched on and there is no interferer present, because the Viterbi decoder may be fed overly pessimistic SNR measurements. Well, I've got interference, so that's not an issue.

MLC and BPF will lead to increased CPU utilization, but I don't really notice it. I've got a 2.6 GHz AMD Phenom II quad core - not exactly cutting edge - and typical utilization is about 35% across all cores.

RNZI DRM @ 17675 kHz, 20 Jul 2012

I just noticed something. When displaying UTC time, notice in the screenshot below that the Excalibur GUI is ahead by a month. Local time is fine. Odd.

Anyway, was getting only about 10 dB SNR until around 04:24 UTC signal power suddenly went way up to about what you see now, maxing out at-95 dBm over a -120 dBm noise floor. Reported SNR in Dream ranges from 18-19 dB. It looks good, like Vatican Radio DRM from Sackville.

This is with a new antenna, mind you. I'll have something more to say about that later.

In the last two minutes of the broadcast, something started up around 17685 kHz. Lots of harmonics, some of it caused co-channel interference.

CRI not so hoggish tonight

Signal's kind of weak on 6020 kHz.

And 9570 kHz is almost non-existent, despite being broadcast from the same site. See lower part of the waterfall, or the demod spectrum on the right. Carrier is -115 dBm.

And the Cuba relay on 9580 kHz has some weird low frequency hum / intermodulation distortion. But the Cuban relay has always been sketchy to me, whether it's distortion or severe attenuation. Radio Habana on 6000 kHz has no such issues. Interesting.

REE DRM @ 9630 kHz, 15 Jul 2012

Okay, here we go. You'll see that I'm trying my luck with manual gain of 73 dB, which gets about -60 dB input power in Dream. Still little in the way of success.

There does happen to be a carrier at 9625 kHz. Could be CBC Radio Nord Quebec out of Sackville, could be Radio Fides out of Bolivia. Impossible to tell. Carrier's -110 dBm at most, and it appears that the lower sideband is under the noise floor, at -130 dBm.

There is also a carrier at 9620 kHz, which is REE in double sideband, broadcast from Noblejas Spain. Crowded band.

A neat thing the Excalibur can do is demod two targets in close proximity to each other. So I'm listening to what pieces of REE 9630 kHz I can get via Dream, and REE on 9620 kHz via the receiver GUI. It's a simulcast.

Subjectively speaking, you need at least half of the OFDM sub-carriers to be demod-able over the interleaving period in order to get any kind of audio. From the waterfall it's obvious that isn't happening.

There is still a 9630 kHz carrier, which is probably Radio Aparecida out of Brazil on 10 kW. REE DRM is itself 30 kW out of Costa Rica. Judging by the level of the carrier, the low frequency audio content is probably about the same power as the DRM signal, but it's all a wash. The transmission has an average SNR of only about 10 dB.

Premature DRM transmission on 9630 kHz

I was just looking over the signal environment in advance of REE DRM at 0:00 UTC.

On the lower 5 kHz there's some irritating tac-tac-tac interference, around -130 dBm. But there also appear to be a carrier at 9630 kHz (which should be suppressed for DRM) as well as what could be two pilot tones.

Usually there is an actual double sideband station at 9625 kHz, CBC Radio Nord Quebec, that caused interference problems in the past. It doesn't seem to be broadcasting at the moment, or anymore? Maybe the FM relays that were supposed to take over from Sackville are now online.

Suddenly, someone must have jumped the gun and fired the DRM broadcast.

Which was short-lived, but Dream actually acquired frame sync and was reporting about 10 dB SNR before someone realized their mistake.

The Disco Palace DRM @ 17875 kHz, 15 Jul 2012

Yeah, not happening. I need signal gain, at least +20 dB of it.

But if you look real close, you can see diagonal striping.

Vatican Radio DRM @ 9800 kHz, 15 Jul 2012

There was a thunderstorm that cleared up about 20 minutes before broadcast began. Transmission started out with a marginal 10 dB SNR, good for a few bursts of audio. It is now almost 15 minutes in, and the signal is almost into the interference and flopping around.

Typically, there is diagonal striping pattern, as fading sweeps slowly from one band edge to the other, and usually always in the same direction. My guess is that the rapid zigzag pattern is coming from the interference.

It's now 20:05 UTC:

On the right hand side, interference has totally consumed the signal. The Dream DRM receiver is losing frame sync. Today's transmission is a write-off.

Interference on the waterfall

Some images of what I have to live with, using the AM loop. Might also be useful for later comparison.

In-channel interference when trying to get Vatican Radio DRM, peaking to -120 dBm. But if we zoom out a bit, we can see that this interferer is periodic in the band.

This interference doesn't experience fading, so it's got to be somewhere close. The antenna is only a metre from the receiver, which isn't too far from power and my desktop. There might be some intermodulation thrown in as well, but I really don't know. I can't exactly reposition this antenna away from potential interferers. Re-orienting the plane of the loop doesn't do anything either, but I wouldn't expect 30 dB nulls from this thing anyway.

Below are a couple more snapshots of 40 m, 20 m, and 17 m, all on 500 kHz DDC bandwidth. The fact that the bandwidth of the interference changes may mean there is more than one source. Or not.

Current receiver setup

I purchased a WR-G31DDC Excalibur from Grove Enterprises prior to a price hike. I lacked, and still lack, a proper antenna. I borrowed an AM loop antenna from an unused CD/radio set, attached the provided SMA-Male to BNC-Female adapter, and stuck one end of the antenna into the inner conductor of the BNC side.

I tried attaching the other end to the outer conductor, but all I got was a material reduction in gain with no material reduction in noise. Evening noise floors are typically about -135 dBm to -130 dBm in the 40 m, 20 m, and 17 m amateur bands, and -125 dBm to -120 dBm in the 49 m and 31 m broadcast bands. That sounds pretty good in that it's not amplifying noise, but then again there's no signal gain to speak of, and the receiver does pick up strong, -110 dBm interferers.

For what amounts to a random piece of wire, and a very short antenna in wavelength terms, it does pretty well for transmissions where I'm in the target area. Today was a particularly good day for Sackville's DRM relay of Vatican Radio, and for most of the broadcast I was getting -95 dBm across 10 kHz, although partially in the teeth of a -120 dBm interferer. Still, that's 25 dB SNR, right? Good enough for 64-QAM, and I had very few drop outs.

And of course, CRI is impossible not to hear.

What I've been dying to receive is The Disco Palace on 17875 kHz. I'm in the target area (i.e. North America), and yet I've only ever gotten sporadic audio when SNR is on the border line of 13 dB. Most times, it's below 10 dB. I figure I'm going to need another antenna to even have a shot at a listenable experience. That should be remedied soon.

China Radio International bit of a bandwidth hog

And by bandwidth, I mean spectrum.

I've noticed this for a while, but didn't find much mention of it from Google, besides a couple German forum threads [1][2].

Basically, China Radio International, as relayed via Cerrik Albania, uses 20 kHz of bandwidth, double sideband.

This is from the English broadcast, at 9570 kHz. At +/- 5 kHz there what look like carriers, likely corresponding to Radio Tupi (Brazil) and Radio Mediterranee International (Morocco), respectively. Any attempt at improved audio quality is thus defeated, with the carriers manifesting as annoying 5 kHz whistles.

Also, not all shortwave receivers can do 20 kHz demod. So what's the point? It seems kind of asinine.

At the same time, CRI is going strong on 6020 kHz, and here the clobbering of other stations is much more apparent.

At 6030 kHz is Radio Marti (USA). I guess the Chinese are doing the Cubans a favour by clobbering it?

I wish I had saved the spectrum from when CRI was relayed via Sackville Canada, but Sackville no longer relays CRI and barring a miracle, will be shut down entirely.