PI4 beacons and the"wrong" VHF digital modes

For discussion of MGM (digimodes) on 4 m, e.g. PSK, FSK

PI4 beacons and the"wrong" VHF digital modes

Postby GM4FVM » Sat 18 May 2013, 13:10

There is a piece in the June 2013 edition of the RSGB "RadCom" magazine about using the "wrong" data modes on VHF.

It is illustrated with a sectrogram of the PI7CIS beacon showing aircraft Doppler shift. It points out that WSPR (the one he chose, but the same principle applies to most other frequency shift digital modes) often fails to decode a signal due to Doppler shift. I think we knew this. Apparently this risks the operator giving up, moving to SSB or RTTY "that does work" and being lost to these modes. Shocking.

I think that this represents a total misunderstanding of what most 4m WSPR operators do. I am not knocking G4JNT and all the good work that he has done, I just think he has missed the point here.

He quotes one operator "the problem is that those who do run WSPR leap for joy when they get a decode outside their normal crop, not realising that they are missing over 95% of those signals". Yes, and FM operators run the same frightening risks every day. We must submit ourselves to the Squelch Inspection.

Firstly, this is not a problem for those of us who do run VHF WPSR, so who cares? We work to get decodes DESPITE the frequency shifts, we learn about paths and openings, and we do get decodes at times when for operational reasons we cannot use SSB or CW (or even that old dinosaur RTTY). It isn't what WSPR misses that matters, it is what it does decode - for instance it is on while I write this and cannot listen, and I have just received a spot in Austria on 6m. Why argue that I could have heard six countries when I cannot listen, when I have decoded one while doing something else?

Secondly, if I did not decode 95% of the signals arriving here I would not get more than a couple of decodes a day, whereas I get far more than that. And I run the WPSR because of the random spots - like the 4m one into Germany in February. It does not bother me that I lose some tropo spots due to frequency shift if I get some via Sporadic E to compensate. That applies even if I have to run WSPR for weeks only to conclude that the path from here into Germany is difficult (but not impossible) on 70MHz during the winter.

Who said this is supposed to be easy? What would be the point of studying propagation if all the reception reports were perfect. Of course sometimes we fail; that is what we are studying.

It is actually BECAUSE these modes are affected by aircraft Doppler shift that I experiment with them, maintaining long paths over long periods. I do not need instant gratification to do what I do. How anyone can think that those of us who dare to use the band on the days outside the contests or the sporadic E season have not got some purpose in mind? We are not put off by the fact that it is difficult. This study simply cannot be done by "devoting the same amount of calling effort to, say SSB or ... JT65B", as anyone who uses WSPR could point out - it works automatically. Automatic SSB, now that would be something.

The same article suggests that Bo's, OZ2M, PI4 Beacon Mode "will run a poor second to JT4 on beacons, despite being designed specifically for use there". Yawn. "Until the decoder software reaches the same proven level of capability" he says - but how do you do that without starting somewhere? How can anything be proven until it is tried? If we all took this view, we need not get out of bed in the morning.

I think that Bo and his team are doing a great job, working in a new and challenging area. It is churlish to taunt them at this early stage. Please keep up development of PI4.

I see this negative attitude a lot on this site too. People go out of their way, and presumably spend hours doing it, to write long deprecating pieces about such-and-such proposed rig, or certain modes. They clearly dislike the idea that somebody somewhere is doing something, or carrying out research or whatever. The same old "SSB is good enough for everyone", or whatever your particular bonkers passion might happen to be, should rule over everybody else's.

Why shouldn't some of us do something different?

He is entitled to say what he thinks. But then surely so am I, the difference being that I encourage everybody to experiment rather than accept the wide-spread uniformity so beloved of the equipment retailers and the self-appointed grandees of the hobby.

I thought this was an experimental hobby.

Jim
GM4FVM
GM4FVM
 
Posts: 173
Joined: Fri 12 Mar 2010, 15:33
Location and locator: BERWICKSHIRE IO85WU

Re: PI4 beacons and the"wrong" VHF digital modes

Postby M0NKA » Sun 19 May 2013, 19:24

Hi Jim,

First, i was one of those who expressed negative opinion of that 70Mhz rig, but the main reason was the price. I design embedded gadgets for a living, been to the Shenzhen market to buy parts, sat next to the Chinese girls soldering boards by hand, etc. In my experience price is based on BOM and R&D expenses, not what the market can bare, or what is the price of a CB rig on Ebay.

I love WSPR too, been through the source code many times, have working TX routines coded in C from scratch, coded the wsprorg.net POST routines in low level C via lwIP, also spent some time in converting the decoder from Fortran to C(still work in progress). So the limitations of WSPR are clear to me, but still it is an amazing mode. The best part for me is also that is automatic, so i can leave it to run now, and analyze the results later. I have a busy lifestyle with kids, work projects, paying bills, etc.

Also WSPR was the best mode that allowed me to design and build my homebrew rig. If i had to use RSBG contests for testing and adjustments - it would have taken many years.

By careful searching the Net, it would be obvious that experimentation in Ham radio is alive and thriving. RSGB magazine unfortunately is not one of those places.

73, Chris
M0NKA
 
Posts: 69
Joined: Wed 31 Aug 2011, 12:09
Location and locator: IO92AM

Re: PI4 beacons and the"wrong" VHF digital modes

Postby GM4FVM » Sun 19 May 2013, 20:42

Chris

Well said.

Your comments about price and marketability are not what I was droning on about. The price is always a subjective judgement and everybody has a view.

Fortran! I was programmer back in the 1970s when the central processor occupied an entire floor of a large building and had to be craned in through the roof. Even then we were trying to get away from Fortran, but it certainly had its uses.

WSPR is not perfect but it does things no other mode can do.

Best wishes
Jim
P.S. WSPR is great for antenna experiments too.
GM4FVM
 
Posts: 173
Joined: Fri 12 Mar 2010, 15:33
Location and locator: BERWICKSHIRE IO85WU

Re: PI4 beacons and the"wrong" VHF digital modes

Postby OZ2M » Wed 22 May 2013, 06:30

Hi all

Ham radio is a hobby. We do things because we enjoy them not because they are easy. To quote JFK: "... We choose to go to the moon in this decade and do the other things, not because they are easy, but because they are hard ..." I mean if we could all make a hole-in-one every time we hit a golf ball it would be uninteresting to play golf. Hobbies are the things we do because they are hard/difficult yet we enjoy them. In this day an age buying a PC and an Internet connection can take us around the globe without investing in rigs, antenna and big PAs.

WSPR may be less fit on VHF and above. But still some find it challenging to do it. What is the problem? If the purpose is to see what X can do there is only one way to find out.

When it comes the comments about PI4 vs JT4 I think it should have been PI-RX vs WSJT instead. Furthermore, does is look like the evaluation of PI-RX was done with the first version because the article mentions 50-432 MHz. The 28 MHz beacon was put on air 18 April. Until v. 0.9.0.10 there was a major bug in the decoder. An integer was wrongly signed :-( This caused a lot of gibberish decodes. From 30 March to 11 May there was a H/W error in the supply cable to the GPSDO causing it to loose power and get out of sync thus no PI4 to decode at times. The feedback we have received on the new version 0.9.1.0 is very good and we would love even more feedback in wav files to improve the PI-RX decoder even more. Why shouldn't PI-RX be able to be on par with WSJT? The source code is available so we could just have made a 1:1 copy but we chose to take a different approach. We set out of a difficult journey but we will succeed.

PI4 takes about 25 s and not 20 s.

Don't frown upon people who do things - do it upon people who don't.
73
Bo, OZ2M - Looking for DX? Try VUSHF DX-peditions www.rudius.net/dxp
OZ2M
Site Admin
 
Posts: 114
Joined: Mon 24 May 2004, 18:41
Location and locator: Herlev 10 km west of Copenhagen, JO65FR

Re: PI4 beacons and the"wrong" VHF digital modes

Postby G0NBD » Thu 30 May 2013, 23:37

Interesting thread,

I must admit one of the best things about the on-off keyed Opera beacon/qso mode , are all the reasons it will not work :) but its actually immune to Doppler spread and has afc routines to enable 'normal' sound cards and ham kit to be used .

Did anyone realize , that if standard 45 baud 2 tone rtty is is reduced in baud rate to the same as wspr , then it is only 6 dB short of the wspr system and that uses 4 mfsk ?

For those who would like to try the qso version of the opera system, the last version with the qso mode, 120 second tx time , -20 dB s/n and 15 chrs , is located at :-
https://dl.dropbox.com/u/15990181/Opera ... 20Beta.zip
Ignore the requests to up-grade ! and have fun - any am/fm pmr set that can be keyed will send the mode ! fsk can also be used , as long as the 'mark' tone is inside the decoder pass band .

Otherwise ,download the latest version , the ROS data mode is also suited to VHF work and can be seen to be used to good advantage on 50 Mhz , being a free running qso mode with no time locking or tx time limitations , the latest versions are quite refined , after 3 years of development .( yes its that long !) There are no 4 meter spots on the psk-reporter , but these are showing 31/5 for 6 meters , im sure similar results could be obtained on 4 , ideal for /P as no web time locking is required for Opera or Ros

73-Graham
G0NBD

EA5AWF 4X4-2175 6m ROS 3298 kms 18:41:58
EC5CFV 4X4-2175 6m ROS 3260 kms 18:57:52
RA3ZSE IV3LWZ 6m ROS 1892 kms 18:42:37
EA5AWF G4MDH 6m ROS 1526 kms 16:53:10
EA5HCJ DK6SQ 6m ROS 1479 kms 17:11:03
30LAJ19 DK6SQ 6m ROS 1479 kms 17:10:24
EA5AWF IT9HZC 6m ROS 1397 kms 17:18:49
F4EJW EA5HCJ 6m ROS 1217 kms 17:19:40
EA5HCJ F4EJW 6m ROS 1217 kms 17:19:28
EA5AWF F4EJW 6m ROS 1202 kms 18:17:39
G0NBD
 
Posts: 156
Joined: Tue 10 Jan 2006, 21:04

Re: PI4 beacons and the"wrong" VHF digital modes

Postby GM4FVM » Fri 31 May 2013, 14:37

The best I had managed on WSPR on 4m was 1115 km, but this week I increased it to 1691km.

I know that others have done better.

Yes, I have had QSOs of double that distance on SSB, but then nobody at that range is trying WSPR yet.

Not bad for the "wrong" mode.

In practical terms, the only snags I have with Doppler are with relatively local stations. I am not really so bothered about that ...

Jim
GM4FVM
GM4FVM
 
Posts: 173
Joined: Fri 12 Mar 2010, 15:33
Location and locator: BERWICKSHIRE IO85WU

Re: PI4 beacons and the"wrong" VHF digital modes

Postby OZ2M » Sat 1 Jun 2013, 19:01

Hi

I don't want to spoil the party. But today OZ7IGY was again decoded via aurora. Is aurora multipath? Guess so.

How do JT65 and JT4 perform via aurora?
73
Bo, OZ2M - Looking for DX? Try VUSHF DX-peditions www.rudius.net/dxp
OZ2M
Site Admin
 
Posts: 114
Joined: Mon 24 May 2004, 18:41
Location and locator: Herlev 10 km west of Copenhagen, JO65FR

Re: PI4 beacons and the"wrong" VHF digital modes

Postby GM4FVM » Sun 2 Jun 2013, 21:29

Tee hee!

Well done. I know PI-RX is worth developing. I wish that I had more time to listen now.

I was reading the web site of a German station listening to IGY on 2m using aircraft scatter. I think that it was over 500km. So PI-RX likes aircraft reflections as well as aurora.

I cannot say that I have increased my 1691km with WSPR on 4m, but I did it today on 10mW.

The Wrong Mode?

Jim
GM4FVM
GM4FVM
 
Posts: 173
Joined: Fri 12 Mar 2010, 15:33
Location and locator: BERWICKSHIRE IO85WU

Re: PI4 beacons and the"wrong" VHF digital modes

Postby GI0GDP » Thu 29 Sep 2016, 08:24

Not really a reply but just to say at present GB3CFG is running PI4 but only on the SE beam at present
regards
Geoff
GI0GDP
GI0GDP
 
Posts: 140
Joined: Tue 20 Dec 2005, 10:20
Location and locator: Carrickfergus, IO74CR


Return to Digital modes

Who is online

Users browsing this forum: No registered users and 1 guest

cron