Porting DSD to Hardware

[jcardani]

Here's a novel idea. How about starting a working group to design a hardware solution around Unitrunker and DSD? Obviously we would need experts in circuit design, DSP hardware design, software development, and board building/testing.

I would envision a black box that would interface with a computer via ethernet IP address so it can be controlled by and viewed in any browser (similar to a Barix Instreamer). Inputs to the black box would include a discriminator (or IF) from a scanner and outputs would be RS232 or USB for scanner/receiver control.

The hardware solution would read and display legacy trunking control formats such as Motorola 3600, EDACS 9600 and 4800, LTR, Passport, Multi-Net, P25 trunking, as well as DMR trunking such as Nexedge. Also, it would read specific talkgroup and unit id data on conventional P-25, conventional DMR (MOTOTRBO and NEXEDGE) and trunked P-25 and trunked DMR voice channels (NEXEDGE).

P25 and DMR voice decode would be off limits unless we can use the AMBE DVSI chips, but that still leaves no solution for IMBE. Unless somehow DVSI's DSP software can be used.

Designing a hardware solution would dramatically improve data decoding accuracy and be a much simpler solution for the average user.

Do we have the expertise and willingness to design such a beast? I would welcome a discussion on the positives and negatives about this idea.

thanks,

 

[IanWraith]

Hello Joe

You might be easier going for ready built hardware like the Hawkboard ..

http://www.farnell.com/datasheets/516890.pdf

Its a standalone computer that can run Linux and they are cheap with mine costing around 80 UK Pounds and usually electronics are even cheaper on your side of the Atlantic. It has sound capability so you may be able to compile DSD directly on to it.

Regards

Ian

 

[gary123]

voice decoding could easilly be done by simply looping the audio through a UIniden bci card from a 785, 796 or 250 scanner.

 

[grosminet]

Alinco DJX-11

Alinco DJX-11 should be a good solution with IQ and discriminator output ready .

don't forget european market with TETRA and TETRAPOL and mpt1327 trunked network

 

[gary123]

Im actually surprised no one has a scanner on the market that would allow the incorporation of additional decode modes. That way the scanning community could write programs to support "new" modes. Many of the digital modes in use are using the same AMBE/IMBE vocoders, its just a matter of supplying them with the "decoded" data blocks.

 

[gmclam]

I am actually able to design, prototype & bold a device like this. While it is a "nice to have", it's not something I can dedicate a lot of time to right now. I see this as an audio decode module/function, main CPU and TCP/IP interface.

Im actually surprised no one has a scanner on the market that would allow the incorporation of additional decode modes. That way the scanning community could write programs to support "new" modes. Many of the digital modes in use are using the same AMBE/IMBE vocoders, its just a matter of supplying them with the "decoded" data blocks.

Consider the royalty costs for digital scanners, and I can see exactly why this is not available. Manufacturers look at how much investment will yield how much profit. I don't think users will pay enough to make it a worthwhile investment.

 

[gary123]

Consider the royalty costs for digital scanners, and I can see exactly why this is not available. Manufacturers look at how much investment will yield how much profit. I don't think users will pay enough to make it a worthwhile investment.

True. however it wouldnt be too hard to take a product like the Uniden 996t (xt) add in a bit of extra memory and create a sub routine in the scanner operating sustem that woud allow the scanner to operate in a "user selected decode mode" . My gesss is that the manufactured would not want to release the code nessessary to drive the scanner in such a "user selected" mode.

I think this is off topic for this thread maybe I should start a new thread on this.

A prototype would be nice. I would love a stand alone device that if nothing else would display parameters such as TG, Source Id, Destination Id, time slot and so forth. DSD is a great program and I use it all the time. Even if such a device just took raw discriminator audio and did everything else it would be great.

 

[Mike_G_D]

Fwi -

Awhile back I started this thread which may have relevance to this subject: http://forums.radioreference.com/tr...digital-demodulation-hardware-discussion.html.

-Mike

 

[ronenp]

I dont understand something

Why building additional Hardware box if it inteface a PC and there are solutions as software like the DSD and others that decode anyway on a PC without any additional Hardware
If you would say that there is a Hardware solution that get Discriminator input or BaseBand and give out a pure oudio without neeed for any PC that would be another thing but as long as PC involved I prefer letting the software do the job as much as posiible and not involveing additional hardware ....

also software changes are much more easy to mede rather then a Hardware changes (unless the hardware is FPGA )

Ronen - 4Z4ZQ
Ronen Pinchooks (4Z4ZQ) WebSite

 

[jcardani]

The reason to develop this in hardware is that it would not be dependent on PC processing and various soundcard configurations. We have seen that certain soundcards work better with DSD than others.

The PC is only there to interface with the hardware to 1) display trunking and conventional data to a monitor and 2) change parameters in the hardware. The processing will totally be done in the hardware not the PC. This way the decoding can be done at the same standard.

However I can see from the posts here that there is little interest in forming a working group.

 

[gary123]

However I can see from the posts here that there is little interest in forming a working group.

Not true.

Im not a programmer but do have some background in electronics. I use a dedicated laptop for DSD and find its often inconvienient to use it as my modile laptop is windows based. A hardware device would be highly desirable.

breadboarding a device would not be too difficult as there are numerous freeware programs for pc board design. Audio output should fall into the same category as there are lots of freeware MP3 decoder routines and chips that will take the recovered digital audio and process it to analog. Im sure that displaying the output info may not be a challenge as there are no doubt a lot of generic LCD displays, chips and drivers available.

 

[mancow]

I think it's a great idea.

 

[wwhitby]

However I can see from the posts here that there is little interest in forming a working group.

Joe:

I think its a fantastic idea! Remember back in the '90s with the equipment that Optoelectronics and later AOR (their P-25 demodulator) used to sell? I see your concept as a logical progression from those pieces of hardware.

My big problem with DSD is that I don't have a computer that I can dedicate to Linux, nor the money to buy an old "cheapie" laptop. I tried the persistant Ubuntu thumb drive route, but only one of my computers will boot from a USB device, and that one would crash with a persistant thumb drive. Even if the "black box" only ran DSD to demodulate the audio, it would be a huge help since it would eliminate the use of a second computer.

Us radio geeks (I use the term proudly) used to build our own stuff years ago. Maybe its time to bring that back!

Warren

 

[ronenp]

I stil think that if a pc is needed it is not neccessary to add additional hardware and that the PC an do the work
I dont think that the sound card is a problerm
There are programs such Power SDR that makes a SDR Radio and it is much more complicated tghen just decoding Single Modulation and they work well with almost any Soundcard .....

But if you still plan to make it with hardware i would like it to decode as much systems as posible specialy motoTRBO ,, NXDN IDAS and D-star
Ronen - 4Z4ZQ
Ronen Pinchooks (4Z4ZQ) WebSite

 

[jcardani]

OK, would there be more interest to do this entirely in software using the soundcard rather than hardware?

I am looking for experienced coders and hardware experts to form a working group. I do not have the expertise alone to do this.

If software is used, we cannot use the software vocoder algorithms like DSD does since there is intel property issues. We can however interface to the hardware AMBE chip that DVSI sells. The software would display TGID, Unit ID, etc. for DMR trunked control, DMR trunked voice, and DMR conventional voice channels.

 

[jcardani]

Another option is to convince Rick to add DMR trunking and DMR conventional decoding to Unitrunker. I know Rick can be real busy - so maybe if we are real nice to him

 

[ronenp]

Why there is an Intel properiaty ? What the diffrence between the DSD and a software we will write ?
I miss something
Unfortenetly Im not a programmer ider so i cant help you ... but on Hardware ((up to but not include ) FPGA programming) i think i can assist with my spare time ..

 

[fourthhorseman]

What the diffrence between the DSD and a software we will write ?

Hopefully This app will be easier to set up an get running,,
somthing in a windows flavor would be nice,,

would this project accept -donations- to maybe get the ball rolling?
i wouldnt mind dropping 20-25 bucks into a project that could be so
much fun as an end user,,

im sure others would be open to it also,,

 

[blantonl]

Im actually surprised no one has a scanner on the market that would allow the incorporation of additional decode modes. That way the scanning community could write programs to support "new" modes. Many of the digital modes in use are using the same AMBE/IMBE vocoders, its just a matter of supplying them with the "decoded" data blocks.

Unfortunately, it's not that easy.

Most of the DVSI license agreements with a manufacturer are application specific. You aren't licensing the vocoder itself, you are licensing your application's use of the vocoder.

 

[jcardani]

Now this looks interesting for the project!


New product from DVSI:

The Digital Voice Systems, Inc. (DVSI) USB-3000™ is the perfect combination of the AMBE-3000™ Vocoder Chip with a USB interface. Simply connect the USB-3000™ to a Windows based PC's USB interface and get direct access to DVSI's AMBE-3000™ Vocoder Chip. This connection allows users to configure vocoder rate and options then encode and decode files, or process real time speech.

DVSI Products AMBE-3000™-HDK