Pages: [1]   Go Down

Author Topic: The 20th...  (Read 7269 times)

K7PEH

  • Member
  • Posts: 1146
The 20th...
« on: February 07, 2007, 07:38:42 PM »

Unless I have counted incorrectly this is the 20th topic posting on this forum since it was created in 2005.  Wow, not much activity.

The other day I was at Border's Bookstore and there was a book on Software Defined Radio that I was looking through.  I didn't buy it though I was tempted.  Maybe the $100 price pushed me away a bit.  This was not a ham radio oriented book but rather a collection of papers on the state-of-the-art and I am sorry, I do not remember the title or editor (aka author).

But, I got the impression that there is a ton of work being done in SDR and I am wondering how far this has gone.  So, let me ask a few questions?

What can SDR not do?  I mean, I am assuming that with SDR you at least need to provide some external (external to the computer and software that is) RF amp and possibly detector circuit although I am not so sure about the detector.  Does DSP handle RF signals and if so where does it end, that is, at what frequencies (VHF ?) is the computer and software not fast enough.

Most of the ham radio oriented SDR that I have seen or read about is actual hardware with some embedded processors and chips and possibly using audio and graphics on a computer for interaction.  But, this book led me to believe that SDR is heading to do an entire radio, with the possible exception of RF preamplifiers, in the computer.

Any comments or answers to my questions?

Phil, K7PEH
Logged

N2AXZ

  • Member
  • Posts: 90
The 20th...
« Reply #1 on: February 07, 2007, 10:14:05 PM »

Hi Phil:

The usual practice with a SDR is to either digitize the signal at some low IF after the incoming RF is put through a standard analog mixer, or digitize and downconvert the incoming signal in one fell swoop through undersampling.   Intentionally undersampling a signal is a clever way of downconverting without the use of a local oscillator --- the IF is generated through the process of aliasing.  Regarding the highest frequency at which DSP devices can operate --- I don't really know what the state of the art is in this area.  Texas Instruments has typically been the leader in the DSP market for a long time, so I would suggest visiting their website to see how fast some of their top-end DSP chips will run.

Hope this helps!

David, N2AXZ
Logged

N9DG

  • Member
  • Posts: 365
The 20th...
« Reply #2 on: February 09, 2007, 07:06:50 PM »

K7PEH: "Unless I have counted incorrectly this is the 20th topic posting on this forum since it was created in 2005. Wow, not much activity."

There's not much activity here on eHam but be assured that there is LOT of stuff going on in amateur radio circles. You just need to look in the right places for it:

http://www.hamsdr.com (~3800 members).

http://hpsdr.org/ (has active discussion forum).

These are some of the discussion forums on Yahoo.com:

SDR-IQ (~350 members).
soft_radio (~1050 members).
softrock40 (~2250 members).

The FlexRadio Systems webpage also has a link to a *very* active forum as well.

And finally this previously posted link is a very comprehensive list of what is all going on out there for SDR:

http://f4dan.free.fr/sdr_eng.html


K7PEH: "What can SDR not do?"

Depends on its architecture is the simple answer. A broad generality is that the closer that the A to D sampling of the RF signal occurs then fewer restriction the SDR system will have. It will take more software obviously, but the overall limitations will be less. Obviously there will always be some hardware needed. And much of that will be manual controls should you need or even want them.


K7PEH: "Does DSP handle RF signals and if so where does it end, that is, at what frequencies (VHF ?) is the computer and software not fast enough."

For the last couple years the CPU power of the host PC is not the limiting factor of the PC based system. One of the neat things about the PC based approaches is that you can mix and match hardware and software to a pretty high degree quite easily. This is true even though much of the software is optimized for a particular piece of HW. There are two basic A to D / sampling approaches being commonly used today:

1. A relatively narrow (192 kHz or less) passband is sampled at up to 192 kHz and at either 16 or 24 bits. These are generally the PC sound card based systems.

2. A broad band approach that samples the entire HF spectrum at 65 MHz and 14 bits or so. These are the radio specific hardware based system that operate independently of the PC sound card, at least for the RX A to D sampling side of things.

They each have their own advantages and disadvantages. And there is also some degree of "blurring" between these descriptions of the two basic approaches.

I'm convinced that amateur radio SDR is as big or even bigger deal for the coming change in ham radio than either SSB or AM were in their day.

Logged

W5CPT

  • Member
  • Posts: 821
The 20th...
« Reply #3 on: February 10, 2007, 07:20:09 AM »

I too have been watching this forum. I have started to consider a new radio. While my IC756 still serves me well it is getting a little long in the tooth. I looked at the present selection of SDRs and was impressed with their capabilities and depressed by the limited selection. I remember the Kachina and also the TT SDR and considered their demise. The new TT entry into the SDR arena, while not a true SDR, is interesting, but not as capable as some.  I guess I am waiting for some competition to lower costs and drive some new features such as spectrum displays and output envelope monitoring. (These may be already available and I may have missed them).

I appreciate those who are developing what will be the future of Ham Radio. I am sure the next breakthrough will not be from the big 3, but until they enter the fray, I believe SDRs will the provence of the small suppliers.  I anxiously await the next entry into the market.

Clint - W5CPT

Logged

N9DG

  • Member
  • Posts: 365
The 20th...
« Reply #4 on: February 10, 2007, 09:33:30 AM »

W5CPT: "I remember the Kachina and also the TT SDR and considered their demise."

I've had a Kachina 505 here and do have several Ten Tec Pegasus radios here that I use almost daily. The Kachina KC-505 was a better performing radio in some regards but the Pegasus easily wins the overall battle because of 3rd party software availability. Kachina totally shot themselves in the foot with the way that they approached software development, - both the in-house and 3rd party software. Ten Tec did a little better job at promoting/assisting 3rd party developers but they are nowhere near the same caliber that FlexRadio Systems is. And FWIW the JA radio companies aren't even as good at it as Ten Tec is in most respects.

The Pegasus and Kachina radios are indeed both exclusively software driven and do use DSP for filtering modulation demodulation etc. internally. However they do not truly capture what the promise of what SDR really is. This is true simply because they are just too similar to traditional radios in their design to be so. In the end they are primarily "software controlled" and the radio's internal "definition" is much more limited than radios like the SDR-1000, SDR-14, SDR-IQ or the various Softrocks are.

Also as far as the Pegasus goes it is a subset of the Jupiter. The Jupiter can also operate in "Pegasus emulation mode". I suspect that many Jupiters are indeed being operated this way and the front panel buttons and knobs are rarely used. The Jupiter just offers a choice of knobs and buttons or just software control.


W5CPT: "I guess I am waiting for some competition to lower costs and drive some new features such as spectrum displays and output envelope monitoring."

When I see things like the DMU units for the new Yaesu's priced at nearly $1000 it is clear to me that we will not see any price competitiveness from the JA companies in this arena anytime soon. For that amount of money you can buy the *entire* PC "backend" that a radio like the SDR-1000 uses. BTW that exact same backend PC can also be used with the SDR-14, SDR-IQ, the Softrocks etc. and probably many more into the future. And when that PC is no longer up to date or capable enough you just give it a way and then buy a new one with more power etc. you can keep the RF specific hardware and use it again with the new PC backend. However with a radio brand/model specific accessory like a DMU you are pretty much stuck with "what you bought is what you keep", no HW upgrade paths there at all. And there is no other non-ham radio use for it either.

As far as RX band scopes go the best performing ones that I've seen yet are the PC based PowerSDR, Winradio, and Rocky programs. Whenever I look at the Icom, Ten Tec, and Yaesu implementations of it I have not been impressed with their sensitivity or refresh rates. Never mind not having a way to just "point and click" to tune in what you do see.


W5CPT: "I appreciate those who are developing what will be the future of Ham Radio. I am sure the next breakthrough will not be from the big 3, but until they enter the fray, I believe SDRs will the provence of the small suppliers. I anxiously await the next entry into the market."

I don't think that the "big 3" entering into the fray will happen anytime soon. Their last round of new model introductions have convinced me of that. And I still have a hunch that we are at the beginnings of the next "changing of the guard" for who our radio manufacturers will be.


And to be clear in my last post I should have said that the FlexRadio *email* reflector is very active. The Forum section has fallen out of favor as "the center of the action" for discussions since the email reflector was implemented.
Logged

KG6OMK

  • Member
  • Posts: 107
The 20th...
« Reply #5 on: March 08, 2007, 06:36:18 PM »

"I got the impression that there is a ton of work being done in SDR and I am wondering how far this has gone. So, let me ask a few questions?"

How far has it gotten?  In the commercial and military worlds it will soon replace everything.  Digital cable TV systems are using OFDM with 256-QAM modulation.  One  does not use just a diode and a coil of wire to either modulate or demodulate this.  This "SDR" technology is in something like 50 million homes now.   Same with over the air digital TV.  In each case they digitize a wide band segment of the IF and the rest is software.

If you want an example of audio only on low bands google on "DRM"  (Digital Radio Modiale.)  It's a broadcast mode.

On the military side there is "JTRS" here is an introduction
http://www.defense-update.com/products/j/JTRS.htm
The program is doing well, with good progress.  Not long before ALL military radio is SDR.

On the ham side we don't see much.  Frankly the technology is over the heads of most hams.  But the promise of SDR is that unlike coils, vacuum tubes and transformers software can be sent over the Internet and  duplicated at no cost.  When the functionality of a radio is defined by software a few guys can build it and send it to everyone else to use.  

There is a project called "GNU Radio" that is very active and is showing some results.  People have used it to make GPS receivers and HDTV.
Take a look here
http://www.gnu.org/software/gnuradio/

Recently the ARRL asked for comments on a new digital mode.  http://www.arrl.org/news/stories/2007/02/22/102/?nc=1

Almost certainly this will actually be a family of digital modes with bandwidths going from "psk31-like" to at least a few kilobits per second.  Some of those will only require a sound card and PC but others may require sampling at the IF level.  As soon as you sample IF you are 100% into SDR.  If this ARRL proposal goes forward SDR will be mainstream.

But you are right, "20 postings".  This shows us that the skills and interests of most hams is other then  software


 
Logged
Pages: [1]   Go Up