Reply
 
LinkBack Thread Tools Search this Thread Display Modes
  #1   Report Post  
Old November 28th 03, 03:26 AM
charlesb
 
Posts: n/a
Default PSK baud rates on HF

Recently I have heard references from multiple sources about 9.6kb PSK
activity on HF bands.

What's the deal on this? Is it a wide, multi-stream mode like Q15x25 mode?

It sounds illegal, but lots of things sound illegal if you are not familiar
with the facts. That's me... I am not knowlegable about PSK packet. I do
have a "flexible" modem though, that will allow me to operate PSK packet at
a variety of baud rates... Isn't 1200 baud the limit on 10 meters, and 300
baud lower down?

Somebody straighten me out on this, please... I'm cornfused.

Charles Brabham, N5PVL
Director: USPacket.Net
http://www.uspacket.net




  #2   Report Post  
Old November 28th 03, 12:29 PM
Paul Keinanen
 
Posts: n/a
Default

On Fri, 28 Nov 2003 03:26:01 GMT, "charlesb"
wrote:

Recently I have heard references from multiple sources about 9.6kb PSK
activity on HF bands.


I assume this refers to 9600 bit/s.


What's the deal on this? Is it a wide, multi-stream mode like Q15x25 mode?

It sounds illegal, but lots of things sound illegal if you are not familiar
with the facts. That's me...


A 9600 bit/s QPSK signal fits nicely within the same bandwidth
occupied by an AM phone transmission, with 8PSK (or 8QAM) the signal
would fit into the bandwidth of SSB phone transmissions. If the
bandwidth limit allows AM or SSB, then what is the problem with other
modes with similar bandwidths ?

I am not knowlegable about PSK packet. I do
have a "flexible" modem though, that will allow me to operate PSK packet at
a variety of baud rates... Isn't 1200 baud the limit on 10 meters, and 300
baud lower down?


That is a purely US specific issue.

Anyway, 1200 baud or 1200 symbols/s is quite capable of transmitting
9600 bit/s provided that the SNR is good. It requires that 8 bits are
transmitted within each symbol, i.e. each symbol can have 256 distinct
states. While 256PSK would hardly be practical, 256QAM might work with
high SNR and low phase distortion line of sight paths.

256QAM requires that the receiver must be able to detect more than 20
distinct states in both I and Q direction, thus quite small errors
will spoil the reception. However 64QAM (6 bits/symbol) are widely
used in Europe on the COFDM subcarriers used in digital television.

Paul OH3LWR

  #3   Report Post  
Old November 28th 03, 12:29 PM
Paul Keinanen
 
Posts: n/a
Default

On Fri, 28 Nov 2003 03:26:01 GMT, "charlesb"
wrote:

Recently I have heard references from multiple sources about 9.6kb PSK
activity on HF bands.


I assume this refers to 9600 bit/s.


What's the deal on this? Is it a wide, multi-stream mode like Q15x25 mode?

It sounds illegal, but lots of things sound illegal if you are not familiar
with the facts. That's me...


A 9600 bit/s QPSK signal fits nicely within the same bandwidth
occupied by an AM phone transmission, with 8PSK (or 8QAM) the signal
would fit into the bandwidth of SSB phone transmissions. If the
bandwidth limit allows AM or SSB, then what is the problem with other
modes with similar bandwidths ?

I am not knowlegable about PSK packet. I do
have a "flexible" modem though, that will allow me to operate PSK packet at
a variety of baud rates... Isn't 1200 baud the limit on 10 meters, and 300
baud lower down?


That is a purely US specific issue.

Anyway, 1200 baud or 1200 symbols/s is quite capable of transmitting
9600 bit/s provided that the SNR is good. It requires that 8 bits are
transmitted within each symbol, i.e. each symbol can have 256 distinct
states. While 256PSK would hardly be practical, 256QAM might work with
high SNR and low phase distortion line of sight paths.

256QAM requires that the receiver must be able to detect more than 20
distinct states in both I and Q direction, thus quite small errors
will spoil the reception. However 64QAM (6 bits/symbol) are widely
used in Europe on the COFDM subcarriers used in digital television.

Paul OH3LWR

  #4   Report Post  
Old November 28th 03, 02:17 PM
Dee D. Flint
 
Posts: n/a
Default


"Paul Keinanen" wrote in message
...
On Fri, 28 Nov 2003 03:26:01 GMT, "charlesb"
wrote:

Recently I have heard references from multiple sources about 9.6kb PSK
activity on HF bands.


I assume this refers to 9600 bit/s.


What's the deal on this? Is it a wide, multi-stream mode like Q15x25

mode?

It sounds illegal, but lots of things sound illegal if you are not

familiar
with the facts. That's me...


A 9600 bit/s QPSK signal fits nicely within the same bandwidth
occupied by an AM phone transmission, with 8PSK (or 8QAM) the signal
would fit into the bandwidth of SSB phone transmissions. If the
bandwidth limit allows AM or SSB, then what is the problem with other
modes with similar bandwidths ?


Well except that for standard PSK enthusiasts tout it's exceptionally narrow
bandwidth thus allowing more conversations in a given space than even CW.
What point is there in creating a wide digital mode?

Dee D. Flint, N8UZE

  #5   Report Post  
Old November 28th 03, 02:17 PM
Dee D. Flint
 
Posts: n/a
Default


"Paul Keinanen" wrote in message
...
On Fri, 28 Nov 2003 03:26:01 GMT, "charlesb"
wrote:

Recently I have heard references from multiple sources about 9.6kb PSK
activity on HF bands.


I assume this refers to 9600 bit/s.


What's the deal on this? Is it a wide, multi-stream mode like Q15x25

mode?

It sounds illegal, but lots of things sound illegal if you are not

familiar
with the facts. That's me...


A 9600 bit/s QPSK signal fits nicely within the same bandwidth
occupied by an AM phone transmission, with 8PSK (or 8QAM) the signal
would fit into the bandwidth of SSB phone transmissions. If the
bandwidth limit allows AM or SSB, then what is the problem with other
modes with similar bandwidths ?


Well except that for standard PSK enthusiasts tout it's exceptionally narrow
bandwidth thus allowing more conversations in a given space than even CW.
What point is there in creating a wide digital mode?

Dee D. Flint, N8UZE



  #6   Report Post  
Old November 28th 03, 03:17 PM
Gene Storey
 
Posts: n/a
Default

"charlesb" wrote

Isn't 1200 baud the limit on 10 meters, and 300 baud lower down?


This is a common misconception about baud and bits. You are correct
about the baud rate limitations, although the baud rate is what is
known legaly as the symbol rate.

Suppose I design a modem that operates at 50 baud (20 ms symbol rate),
obviously I can use this below 10 meters. Now, let's say I modulate
that rate with 36 carriers of DQPSK. That is, there are 36 carriers
of four phases (two bits per carrier), or 72 bits in 20 ms (50 baud).
There's fifty 20 ms periods in a second, so we have (72 bits x 50), or
3600 bps.

Now obviously we have learned something in the last 20 years, and that
is, that HF is a bad medium, and you can either retry your transmissions
until the band quits, or you can insert some sort of error correction into
the transmission, to prevent retransmission. So let's say we limit the
information to 2400 bps and insert 1200 bps of error correction (FEC),
for a total of 3600 bps.

Suppose each of the 36 carriers is a multiple of 62.5 Hz, then we can
say that 62.5 x 36 is 2250 Hz. But we don't really want to go all the
way down to 0 Hz, so let's say we start at 312 Hz (62.5 x 5) and this
will push the right side out to (62.5 x (5 + 35)) or 2500 Hz.

In summary, 36 carriers takes a lot of bandwidth, but no more than
a standard analog sideband. The fact that you can fit 2400 bps
voice, data, and image into the channel might be worth the complexity.

P.S. I just described the G4GUO modem :-)


  #7   Report Post  
Old November 28th 03, 03:17 PM
Gene Storey
 
Posts: n/a
Default

"charlesb" wrote

Isn't 1200 baud the limit on 10 meters, and 300 baud lower down?


This is a common misconception about baud and bits. You are correct
about the baud rate limitations, although the baud rate is what is
known legaly as the symbol rate.

Suppose I design a modem that operates at 50 baud (20 ms symbol rate),
obviously I can use this below 10 meters. Now, let's say I modulate
that rate with 36 carriers of DQPSK. That is, there are 36 carriers
of four phases (two bits per carrier), or 72 bits in 20 ms (50 baud).
There's fifty 20 ms periods in a second, so we have (72 bits x 50), or
3600 bps.

Now obviously we have learned something in the last 20 years, and that
is, that HF is a bad medium, and you can either retry your transmissions
until the band quits, or you can insert some sort of error correction into
the transmission, to prevent retransmission. So let's say we limit the
information to 2400 bps and insert 1200 bps of error correction (FEC),
for a total of 3600 bps.

Suppose each of the 36 carriers is a multiple of 62.5 Hz, then we can
say that 62.5 x 36 is 2250 Hz. But we don't really want to go all the
way down to 0 Hz, so let's say we start at 312 Hz (62.5 x 5) and this
will push the right side out to (62.5 x (5 + 35)) or 2500 Hz.

In summary, 36 carriers takes a lot of bandwidth, but no more than
a standard analog sideband. The fact that you can fit 2400 bps
voice, data, and image into the channel might be worth the complexity.

P.S. I just described the G4GUO modem :-)


  #8   Report Post  
Old November 28th 03, 03:28 PM
Gene Storey
 
Posts: n/a
Default

"Dee D. Flint" wrote

Well except that for standard PSK enthusiasts tout it's exceptionally narrow
bandwidth thus allowing more conversations in a given space than even CW.
What point is there in creating a wide digital mode?

Dee D. Flint, N8UZE


Single channel PSK has nothing in common with high bandwidth information
transport. What one group touts may be exactly opposite of what another
group touts. Sometimes 31 bps at 31 baud is good enough, other times
3600 bps at 50 baud is what is needed.

The point in creating a wide digital mode, is to increase throughput per second.



  #9   Report Post  
Old November 28th 03, 03:28 PM
Gene Storey
 
Posts: n/a
Default

"Dee D. Flint" wrote

Well except that for standard PSK enthusiasts tout it's exceptionally narrow
bandwidth thus allowing more conversations in a given space than even CW.
What point is there in creating a wide digital mode?

Dee D. Flint, N8UZE


Single channel PSK has nothing in common with high bandwidth information
transport. What one group touts may be exactly opposite of what another
group touts. Sometimes 31 bps at 31 baud is good enough, other times
3600 bps at 50 baud is what is needed.

The point in creating a wide digital mode, is to increase throughput per second.



  #10   Report Post  
Old November 29th 03, 01:14 AM
charlesb
 
Posts: n/a
Default


"Dee D. Flint" wrote in message
gy.com...

What point is there in creating a wide digital mode?


More throughput, error correction for 100% copy, ability to transfer binary
files. ( Q15x25 guys like to send JPEG's back 'n forth, kind of like SSTV,
except the pics come through 100% good. )

Obviously any wide mode is overkill for keyboard QSO's, but there's a whole
world of other things that hams want or need to do, and some of them require
more bandwidth.

There are a lot of pro's and cons related to utilizing the wide modes for
any purpose. I don't think we will have to worry about them becoming so
popular that they will squeeze out everybody else.

Charles Brabham, N5PVL
Director: USPacket.Net
http://www.uspacket.net



Reply
Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
My MHeard list for 07/18/03 50.620 FM 1200 Baud Packe from EM89ag N8XYN Digital 0 July 18th 03 10:46 PM
My MHeard list for 07/18/03 50.620 FM 1200 Baud Packe from EM89ag N8XYN Digital 0 July 18th 03 10:46 PM
Tower rates Bressler Communications Antenna 1 July 18th 03 04:57 AM
6 meter packet, 1200 baud, FM, 50.620 mHZ. N8XYN Digital 0 July 5th 03 09:17 PM
6 meter packet, 1200 baud, FM, 50.620 mHZ. N8XYN Digital 0 July 5th 03 09:17 PM


All times are GMT +1. The time now is 01:05 AM.

Powered by vBulletin® Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 RadioBanter.
The comments are property of their posters.
 

About Us

"It's about Radio"

 

Copyright © 2017