Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: PSK31 vs CW...  (Read 4382 times)
KA3DNR
Member

Posts: 74




Ignore
« on: October 07, 2009, 12:01:07 PM »

Hello Folks,

I have read somewhere that PSK31 is much more "efficient" than CW. Have any of you made some empirical comparisons to test this out? In my light-work comparison, the weakest signal I hear on PSK31, I could also work on CW. But, what do I know?

Your thoughts? Do you have any juicy PSK31 sites explaining its superiority?

Regards & Thanks,

Marc
KA3DNR
Logged
W4KVW
Member

Posts: 476




Ignore
« Reply #1 on: October 07, 2009, 02:48:02 PM »

I think the PSK signal is EASIER for WEAK signals but I am NOT an EXPERT just from my limited use of both modes.

Clayton
W4KVW
Logged
K7JBQ
Member

Posts: 80




Ignore
« Reply #2 on: October 07, 2009, 08:10:59 PM »

I use both CW and PSK on a regular basis.

For really weak signal work, CW wins, assuming the processor between the ears is up to snuff.

Stuff I can hear well enough to work on CW, often times results in garbled print on PSK.

73,
Bill
Logged
N3OX
Member

Posts: 8853


WWW

Ignore
« Reply #3 on: October 07, 2009, 10:28:10 PM »

"Stuff I can hear well enough to work on CW, often times results in garbled print on PSK. "

That's also my experience.

My quick Googling suggests that people argue a couple of things:

1) Phase shift keying is better than amplitude shift keying.

2) The bandwidth appropriate for of a CW signal is the width of commonly used CW filters so that the bandwidth of a CW signal is perhaps 10-15 times greater than that of a PSK31 signal.   The increased power spectral density gives a similar boost going from CW to PSK as you get from going from SSB to CW.

I think #1 is absolutely true for computer decoding.  Amplitude shift digital modes are very hard to decode in the presence of noise and interference by comparison.  It's hard to tell "on" and "off" by amplitude when one second's "on" state is 15dB louder than the last second's "on" state.  

#2 isn't really true and I can't make a very rigorous argument at this late hour, but I think there are a lot of people using telecommunications theory without really stating their assumptions well.

It's certainly not the case that a CW signal is spread evenly over a 500Hz or 300Hz filter bandwidth.  Far from it.

Both 1 and 2 fall to an additional problem.  On the "receiving" end... and by that I mean after the radio has done its thing, the CW signals are further processed by an amazing, adaptive, predictive pattern recognizing filter and decoding software.

In the "hardware" of that filter you've got ears and a brain that are able to focus in a trainable way on a narrow segment of received audio.  You can have a few CW signals in your passband and choose to listen to each one and ignore the others.  You can choose to listen to narrow your filter bandwidth down to listen to the coherent switched tones and not the noise.   It's not clear to me how to take that into account in this argument.

And in the "software" end, we can PREDICT that soon there's very important information coming after "RST is" and we're primed and ready to receive that very difficult to hear "349 349 349" from JA7xxx on 160m.

We know that after "CQ CQ DE" comes some other very important information.  

Wouldn't it be nice if you could tell your PSK software to LISTEN HARDER at the right time so your screen didn't look like:

CQ CQ DX CQ DX de V#(gA8FA F(30A
CQ CQ DX C) DX de V8#@) * F)*#@$
CQ CQ DX CQ DX de #$)_0 )*V 3281
PSE K K

?

We don't have to be content with what we actually hear.  We can take what we actually hear and ask ourselves if it was probably true, and look for some sense in it.  We're not going to EVER decode punctuation in a callsign, for example.

Even if it sounds like it, you're not going to decode

CQ de W7?

You're going to figure out it was

CQ de W7UTI

or

CQ de W7IMI

Furthermore, the ear-brain filter knows the difference between signal and noise.  This is something of an interesting concept.  As good pattern recognizers, we just intuitively know signal from noise.  We have a complex set of rules that we apply at blazing speed to tell the difference, and we have the disadvantage that we don't know exactly how it works, so we can't tell our computers how to do it.  We're SO good at it that when we get started in radio we might think "why can't a manufacturer make a radio that just gets rid of the noise?!"

Anyway, I think PSK31 is a good weak signal mode.  But asking if it's better than CW may not even be a well formed question, given the nature of the CW decoder.

And I think there may be some other technical issues (related only to engineering questions, not deep into the neuroscience) that people get a little wrong when they estimate how good PSK31 is vs. CW.

73
Dan
Logged

73,
Dan
http://www.n3ox.net

Monkey/silicon cyborg, beeping at rocks since 1995.
KA3DNR
Member

Posts: 74




Ignore
« Reply #4 on: October 08, 2009, 08:52:42 AM »

Excellent Dan!

Thank you, and others, for taking the time to respond.

What you had stated is down the same train of thought I was going through.

Regards,

Marc
KA3DNR
Logged
AA4PB
Member

Posts: 12695




Ignore
« Reply #5 on: October 08, 2009, 11:17:36 AM »

I don't know, but under certain conditions I've copied some PSK31 signals that I couldn't even hear so I seriously doubt that I could have copied them on CW. Of course using the appropriate IF filter and headphones I might have been able to dig them out of the noise.
Logged
5R8GQ
Member

Posts: 203




Ignore
« Reply #6 on: October 08, 2009, 01:50:52 PM »

Guess it depends on what you mean by "efficient".

The Free Dictionary.com: Efficient

a. Acting or producing effectively with a minimum of waste, expense, or unnecessary effort.

b. Exhibiting a high ratio of output to input.

Minimum of expense and electrical energy:
QRP CW rig, lantern battery and simple wire antenna.

Minimum of unnecessary effort:
Point and click on a PSK signal and start typing.
QRP PSK works really well too.

You can make a good argument for each mode.
Logged
G0GQK
Member

Posts: 634




Ignore
« Reply #7 on: October 08, 2009, 02:00:01 PM »

It has been said that PSK31 is comparable to Morse code because of the narrow signal but I don't think there is an advantage. However if Morse transmissions are compared to modes like WSPR where data can be received when a signal cannot be seen or heard, then WSPR will be the better mode.

G0GQK
Logged
KA3DNR
Member

Posts: 74




Ignore
« Reply #8 on: October 08, 2009, 03:47:47 PM »

Hello Mel,

I didn't know WSPR existed; I am intrigued.

Usage is going up in a linear fashion:

http://wsprnet.org/drupal/wsprnet/stats

I need to learn more about this. It appears to be beacon-like communication, not a QSO per se. I am ignorant. I'll go to the forums and start asking questions.

Thanks again Mel!

Regards,

Marc
KA3DNR
Logged
K5MBV
Member

Posts: 265




Ignore
« Reply #9 on: October 15, 2009, 08:04:43 PM »

Marc, you might be better investigating the WSJT modes
that can yield good copy on signals that can not
be heard at all. FSK441 and JT65B for example. One
for meteor scatter and the other for EME. They are
actual communication programs quite a bit different
from WSPR and can copy signals 20dB or so weaker than
CW can.

Ken  K5MBV
Logged
KA3DNR
Member

Posts: 74




Ignore
« Reply #10 on: October 15, 2009, 10:26:08 PM »

Ahh, very good Ken.

I tried WSPR. While it is cool, it is only beacon-beacon communication. I want to converse with others!

I will check out those modes.

Regards,

Marc
Logged
N8UZE
Member

Posts: 1524




Ignore
« Reply #11 on: October 20, 2009, 04:47:30 PM »

Each mode will have its advantages and disadvantages.  Depending on circumstances, one will be "better" or "more efficient" than the other.  There are circumstances where CW is distorted but can be still be copied while PSK will be totally garbled.  On the other hand, PSK signals that can't even be heard by ear can be copyable.

There is NO one best mode.
Logged
AB7E
Member

Posts: 117




Ignore
« Reply #12 on: November 03, 2009, 09:48:55 PM »

I'm surprised no one has discussed efficiency in terms of bandwidth.  CW wins hands down for readability with fading and marginal S/N, but it's pretty difficult for ten different CW QSOs to co-exist in a half KHz portion of the band and it's not unusual at all to be able to do so with PSK31 at reasonable signal strengths.

73,
Dave   AB7E
Logged
N4KZ
Member

Posts: 594




Ignore
« Reply #13 on: November 04, 2009, 10:58:30 AM »

Using G-TOR and MFSK16, I've had solid QSOs with stations on 20 meters that I could not hear by ear. But their text printed reliably on the monitor. Kind of an eerie feeling. As great a mode as CW is, if you can't hear the station, there won't be a contact. But that is not the case with some digital modes. Personally, I've never had a PSK31 QSO with a signal that I couldn't hear in the speaker. I attribute that to the fact that G-TOR was an ARQ -- automatic repeat request -- mode and MFSK16 offers FEC, forward error correction, which is to say it offers redundancy. Both improve throughout and PSK31 isn't in the same league. But it does offer extremely narrow bandwidth which dramatically improves S/N ratio on the copying end.

When operating on 20-meter PSK31, if you will use the 50 HZ digital filter in your rig, you can begin copying all the 10-watt European stations who live in apartments and run indoor antennas. Quite interesting.
73, N4KZ
Logged
WB5JEO
Member

Posts: 805




Ignore
« Reply #14 on: November 05, 2009, 08:35:31 AM »

Clearly, there's more to "efficiency" than immediately leaps to mind. From the perspective of someone trawling for particular stations or areas, PSK31 lends that ability to monitor, even while working another band or mode, a bunch of "channels" visually. That includes making a prearranged contact on a busy day when the agreed CW frequency is plus or minus and you're hunting for each other. And of course technical efficiency isn't everything. Although experienced CW operators feel the same kind of personal contact working a friend whose style they're familiar with, for most operators, phone is attractive because of, for them, more personal contact in natural speech. And, I suspect, a lot of older operators just don't feel as connected on a computer interface. So, "efficiency" also means what's efficient in terms of establishing a particular kind of connection to another person more readily. And efficiency in terms of getting set up for a mode is another kind of efficiency. PSK31 isn't much more intensive in terms of hardware than CW, with free software and cheap or junkbox soundcard connections, and certainly more quick and easy to set up and use PSK than CW for someone who would have to actually learn the code.
Logged
Pages: [1] 2 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!