- SDR DSP Development board
- Posted by ratemonotonic on June 13th, 2008
Hi all ,
I have a requirement for a DSP board for development of SDR
applications. The carrier frequency we are looking at is max 8Khz so
there is no IF required, we can sample data straight in. The boards
that I have found are mostly with ADCs with high sampling rate
( ~56Mhz). I require something with lower sampling rate and more
precision for dynamic range.
I am looking for boards with TI DSP floating point like C67xx,
Can anyone recommend a board with good BSP.
Thanks
rate
- Posted by Vladimir Vassilevsky on June 13th, 2008
ratemonotonic wrote:
8kHz are you sure? Then pretty much any DSP board with audio codec would
do. May be you can even get by an internal ADC of a microcontroller.
You don't need a cannon to kill the flies.
I have TMS5510 EzDSP that I would like to get rid of for half price.
Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com
- Posted by Tauno Voipio on June 13th, 2008
ratemonotonic wrote:
I used a Motorola MC56k development board for
a similar task, with 9600 samples/s.
The board is now obsolete, so there's no help
for you if I give the exact type for you.
--
Tauno Voipio (OH2UG)
- Posted by Chris Maryan on June 13th, 2008
On Jun 13, 10:16 am, ratemonotonic <niladri1...@gmail.com> wrote:
Do you need a DSP board? At that frequency a PC with a good sound card
will suit you just fine.
Chris
- Posted by ratemonotonic on June 14th, 2008
On Jun 13, 3:43 pm, Vladimir Vassilevsky <antispam_bo...@hotmail.com>
wrote:
The problem is that it is a low bit rate radio with sychronisation
timing constrains , hence have to make the receiver coherent.
Also the environment(s) are really noisy so need a lot of processing
gain and equalisation. Thats the reason I am wanting to go for a
DSP with clout for research and development phase and then I can drop
MFLOPs.
I am using a DSP library which has optimised support for C67xx and
ADSP floating point devices.
BR
Rate
- Posted by Vladimir Vassilevsky on June 14th, 2008
ratemonotonic wrote:
If the receiver is coherent, then the PLL acquisition time adds to the
total synchronization time.
Does the ADC response have to be all way down to DC ?
"Really" and "A lot" are not the engineering terms.
Then use a PC with a sound card for R&D.
This is what they call "premature optimization".
Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com
- Posted by ratemonotonic on June 15th, 2008
of
) are due to a totally novel channel/medium. The characteristics
of which I cannot discuss , so apologetically I cannot continue
discussion on that. Was only looking for advise on DSP boards without
the extra bells and whistles they usually come with.
soundcard , i am looking at a software called Portaudio to access the
soundcard.
Also I am looking to construct a front end to connect to the
soundcard.
I am eager to start the embedded side of development because a PC
based simulation hides a lot of embedded issues.
Any advise on a DSP(and ADC doughter card) board with should BSP
should be gratefully accepted. Thanks
BR
rate
- Posted by Vladimir Vassilevsky on June 15th, 2008
ratemonotonic wrote:
The entities should not be multiplied beyond necessity. The standard
windows wave API functions will do; it is all described in MSDN.
Could this be a problem?
So you really do that for enjoyment of the process rather then to get
the result, right?
Hire the consultant.
Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com
- Posted by ratemonotonic on June 16th, 2008
On 15 Jun, 19:37, Vladimir Vassilevsky <antispam_bo...@hotmail.com>
wrote:
Look at port audio it is interesting!
- Posted by Chris Maryan on June 16th, 2008
On Jun 15, 2:06 pm, ratemonotonic <niladri1...@gmail.com> wrote:
Also, if like a sizable portion of the people here you are using
Matlab, the data acquisition toolbox can access a sound card.
Chris
- Posted by ratemonotonic on June 18th, 2008
On Jun 16, 4:35 pm, Chris Maryan <kmar...@gmail.com> wrote:
Thanks chris.