Re: [Discuss-gnuradio] gmsk2 - CRC and whitening

2005-12-02 Thread Robert Cicconetti
Apologies. That is definitely the wrong version. That one just sends out MARK continuously. Try this one.
R C
On 12/2/05, Robert Cicconetti [EMAIL PROTECTED] wrote:
I've got some half working code for an APRS tracker targeting an -8515.
Doesn't have a crc routine built in (didn't get that far). Maybe one of
these days I'll finish it. :) At some point I did have multimon
decoding my test packet, but I can't guarantee that this is the version
that actually worked. :)

As I recall, it's set to use a 3.6864 Mhz oscillator and to generate
1200/2200 baud signals. It implements a basic HDLC NRZI encoding, and a
software DDS. Data comes out PORTC. Running a 10 khz sample clock, the
beast was asleep most of the time, waiting for a counter to fire.
You're welcome to borrow anything you want from it, although I won't
guarantee anything actually works on it.. haven't touched this code in
ages. Rather crufty, although it is heavily commented. :) It will
currently send out packets containing TestingTesting and my call
sign. Remember that if you change the packet, you need to change the
CRC16 at the end.

R C
On 12/1/05, Daniel O'Connor [EMAIL PROTECTED]
 wrote:
On Fri, 2 Dec 2005 09:27, Eric Blossom wrote:  Processor - Atmel AT90S8535 - face turns beat red. =) The AVR processors are very nice.Sort of a minature RISC architecture. You've got 8K of flash, so I don't see any problem at all.I think
 there's a port of GCC that targets it too.There sure is :)http://savannah.nongnu.org/projects/avr-libc

http://www.nongnu.org/avr-libc/The 8535 is pretty crusty though ;)




fsk2.c
Description: Binary data
___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


[Discuss-gnuradio] gnuradio in Solaris - progress report

2005-12-02 Thread cswiger
From the for-what-its-worth-dept (very low priority):

I've been making stabs at getting Gnuradio to compile
on Solaris (in case a 12 processor SunFire shows up ;)
and todays random hacks finally got it to go thru.
Python can now load the modules, but running is another
matter:

# python
Python 2.4.2 (#1, Dec  1 2005, 14:38:20)
[GCC 3.4.5] on sunos5
Type help, copyright, credits or license for more information.
 from gnuradio import gr
 fg = gr.flow_graph()
 src = gr.sig_source_c(8000,gr.GR_SIN_WAVE,440,1,0)
 sink = gr.null_sink(gr.sizeof_gr_complex)
 fg.connect(src,sink)
 fg.start()
gr_vmcircbuf_createfilemapping: createfilemapping is not available
gr_vmcircbuf_sysv_shm: shmget (2): Invalid argument
gr_vmcircbuf_sysv_shm: shmget (2): Invalid argument
terminate called after throwing an instance of 'gr_signal'

Abort (core dumped)



I guess that has something to do with shared memory?

Anyway, if there's no easy solution with existing code
I'm giving up for the time being. My E250 has two 300Mhz
cpus and 512Mb memory. Boost takes over 4 hours to compile.

--Chuck




___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


[Discuss-gnuradio] A quick answer?

2005-12-02 Thread Robert McGwier
I did a bit of google searchinrg but cannot seem to hit on the cause of 
this:


** (python:1778): WARNING **: IPP request failed with status 1030

I am running the usrp FM receive program.  It doesn't seem to impact the 
performance but it is an annoyance. Anyone have a quick suggestion?


Bob

--
Laziness is the number one inspiration for ingenuity.  Guilty as charged!



___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


RE: [Discuss-gnuradio] gmsk error

2005-12-02 Thread Robitaille, Michael

I am in the process of implementing the whitener and will most likely do
CRC16.  For integrity check, I originally implemented of you a parity bit
(yes it's limited to detecting on one bit error but my data load was small -
two byte).

Thanks,
Mike

-Original Message-
From: Robert McGwier [mailto:[EMAIL PROTECTED] 
Sent: Thursday, December 01, 2005 8:30 PM
To: [EMAIL PROTECTED]
Cc: discuss-gnuradio@gnu.org
Subject: Re: [Discuss-gnuradio] gmsk error

We put the whitener and CRC32 in for a reason:

1) clock recovery and other things work better if you do not have long 
runs of 1's or 0's.

2) How will you be able to check end to end integrity without CRC or 
other style checking?

The answer is,  you can live without either but you pay the obvious 
penalties implied by 1 and 2 above.

Bob



[EMAIL PROTECTED] wrote:

Urgg, should have been under this subject

Original Message:
-
From: [EMAIL PROTECTED] [EMAIL PROTECTED]
Date: Thu, 1 Dec 2005 13:53:15 -0500
To: discuss-gnuradio@gnu.org
Subject: [Discuss-gnuradio] (no subject)


On Wed, 30 Nov 2005 15:00:34 -0800, Eric Blossom wrote:
  

The calling sequence for the gaussian filter design changed.
Even if ggmsk-test were fixed to use it, it doesn't work anywhere
nearly as good as the new code.



I would like to use the new code in gmsk2 but I can't implement the CRC32
and the whiten/dewhiten in the microprocessor used in the transmitter.  The
original engineer on this project, who left the company and that I am
replacing, selected a real poor microprocessor.

Do you think that I can use the same basic idea of the gmsk2 framer and
sync whithout the use of the CRC32 and whiten?

Mike


mail2web - Check your email from the web at
http://mail2web.com/ .




___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


mail2web - Check your email from the web at
http://mail2web.com/ .




___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

  



-- 
Laziness is the number one inspiration for ingenuity.  Guilty as charged!


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


RE: [Discuss-gnuradio] gmsk2 - CRC and whitening

2005-12-02 Thread Robitaille, Michael
I will check out to see if I can find the GCC.

The whitening was very easy to implement.

Thanks,
Mike

-Original Message-
From: Eric Blossom [mailto:[EMAIL PROTECTED] 
Sent: Thursday, December 01, 2005 3:58 PM
To: [EMAIL PROTECTED]
Cc: discuss-gnuradio@gnu.org
Subject: Re: [Discuss-gnuradio] gmsk2 - CRC and whitening

On Thu, Dec 01, 2005 at 05:42:36PM -0500, [EMAIL PROTECTED] wrote:
 Good suggestions, will look into implementing CRC16 with hardware and the
 whitening for small payload.
 
 Processor - Atmel AT90S8535 - face turns beat red. =)

The AVR processors are very nice.  Sort of a minature RISC architecture.
You've got 8K of flash, so I don't see any problem at all.  I think
there's a port of GCC that targets it too.

 Looking at the packet.sink code, it looks like it's independant of the CRC
 code.  Is this right?

Yes.  It just checks that it's found the sync pattern and that the
header length was properly detected.

 gmsk2_pkt.py uses delete_head().  I can't find this code, where is it?

gr_msg_queue.{h,cc}

Eric


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


Re: [Discuss-gnuradio] gnuradio in Solaris - progress report

2005-12-02 Thread Lamar Owen
On Friday 02 December 2005 11:04, cswiger wrote:
 From the for-what-its-worth-dept (very low priority):

 I've been making stabs at getting Gnuradio to compile
 on Solaris (in case a 12 processor SunFire shows up ;)

Oh, something like: 
http://www.lamarowen.net/pgallery/photo?photo%5fid=3034 :-)

I have a guy wanting to donate two E10K's with 64 cpus and matching RAM; while 
I have the power, cooling, and other infrastructure, don't really want to 
drive to San Diego for them... :-)

 Anyway, if there's no easy solution with existing code
 I'm giving up for the time being. My E250 has two 300Mhz
 cpus and 512Mb memory. Boost takes over 4 hours to compile.

You do know there is a Linux for that box, right?  That's what Spacely (the 14 
CPU  (400MHz 8MB ECache) E6500 with 14GB RAM) is running; while I haven't 
tried to build the GNUradio stack on it, it should build fairly easily, since 
the Linux that is on it is based on Fedora Core 3 (Aurora; 
www.auroralinux.org). 

I am getting ready to build up either the E5500 or an E3500 (or perhaps a quad 
proc E450) with Solaris 10; I'd be willing to give you an ssh if it would be 
of help.  Solaris 10 should be closer to the Linux API; there is a linux 
personality module for it, at least on i386 hardware.
-- 
Lamar Owen
Director of Information Technology
Pisgah Astronomical Research Institute
1 PARI Drive
Rosman, NC  28772
(828)862-5554
www.pari.edu


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


[Discuss-gnuradio] USRP Update

2005-12-02 Thread Matt Ettus


In this issue
=
- Flex400 Daughterboard Available
- Daughterboard Status
- Antennas
- Datasheets
- Shipping Updates
- Low stocks on TVRX, DBSRX
- Year-end special

Flex400
===

The Flex400 Transceiver daughterboard is now available.  This board
covers 400 to 500 MHz, both transmit and receive.  Output power is  100
mW, and receive noise figure is 3-5 dB.  It has over 20 MHz bandwidth on
both TX and RX, and can operate full duplex as long as you separate the
TX and RX frequencies (TX and RX have separate PLL synthesizers).  There
are 2 antenna ports, one for TX and RX and one for  RX only.  This
allows simple connection to downstream amplifiers and TR switches.

The Flex400 costs $225, and up to 2 of them may be used with a USRP at
a time.  The Flex400 is in stock now and shipping.  Pictures of the
board are available on the ordering page.  Full design files are
available from the Ettus Research download page:

http://www.ettus.com/Download.html



Daughterboard Status


The long awaited Flex900, Flex1200, and Flex2400 (800-1000 MHz,
1200-1300 MHz, and 2300-2500 MHz respectively) transceiver boards are in
 pre-production now, and should be available in January.  Final pricing
has not been set, but is expected in the $225 to $300 range.


Antennas


In addition to the 4 PCB antennas which we have been selling for some
time, we now carry a tri-band vertical whip which covers 400 MHz and
1200 MHz, as well as the ham 2M band.  These antennas are an ideal match
for the Flex400 and Flex1200, and are available for $45.


Datasheets
==

Datasheets for the USRP and the antennas are now available on the Ettus
Research ordering page, along with pictures of all items for sale.  See:

http://www.ettus.com/custom.html

Shipping Updates


In order to streamline the shipping process, all shipments are made on
Wednesdays.  Any order received by 8 AM Pacific time on Wednesday will
be shipped that Wednesday.  Any orders received after that will have to
wait until the next Wednesday.

US customers now have the option of DHL next day service.  This is
still subject to the Wednesday-only shipping schedule.

Current inventory status is available through a link on the ordering 
page.

Low Stocks on TVRX and DBSRX


We currently have only 8 TVRX and 8 DBSRX boards left in stock.
Replacements may take as long as 8 weeks due to long component lead
times, so order soon if you are going to need either of these boards in
the near future.

Year-end special


We still have 4 pre-production models of the USRP left.  The only
difference between these and the current units is that they do not have
the capability to use an external oscillator, and they have 2 blue
wire jumpers on them.  These units are available for $100 off the
normal price of a USRP.  Please email [EMAIL PROTECTED] if you are interested.


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


Re: [Discuss-gnuradio] gnuradio in Solaris - progress report

2005-12-02 Thread Eric Blossom
On Fri, Dec 02, 2005 at 11:04:44AM -0500, cswiger wrote:
 From the for-what-its-worth-dept (very low priority):
 
 I've been making stabs at getting Gnuradio to compile
 on Solaris (in case a 12 processor SunFire shows up ;)
 and todays random hacks finally got it to go thru.
 Python can now load the modules, but running is another
 matter:
 
 # python
 Python 2.4.2 (#1, Dec  1 2005, 14:38:20)
 [GCC 3.4.5] on sunos5
 Type help, copyright, credits or license for more information.
  from gnuradio import gr
  fg = gr.flow_graph()
  src = gr.sig_source_c(8000,gr.GR_SIN_WAVE,440,1,0)
  sink = gr.null_sink(gr.sizeof_gr_complex)
  fg.connect(src,sink)
  fg.start()
 gr_vmcircbuf_createfilemapping: createfilemapping is not available
 gr_vmcircbuf_sysv_shm: shmget (2): Invalid argument
 gr_vmcircbuf_sysv_shm: shmget (2): Invalid argument
 terminate called after throwing an instance of 'gr_signal'
 
 Abort (core dumped)
 

 I guess that has something to do with shared memory?
 
 Anyway, if there's no easy solution with existing code
 I'm giving up for the time being. My E250 has two 300Mhz
 cpus and 512Mb memory. Boost takes over 4 hours to compile.
 
 --Chuck

Hi Chuck,

I belive that you need to use sysctl to increase the default number of
SysV shared segments and/or their maximum size.

You might try 

  $ ipcs -l

to check the current limits.

Does it pass make check ?

It's been a while since I checked, but gnuradio-core has compiled on
Solaris in the past.

Eric


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio