Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: Prev 1 [2] 3 Next   Go Down
  Print  
Author Topic: Any one actually want to talk about SDR and future implementations?  (Read 10707 times)
N3OX
Member

Posts: 8847


WWW

Ignore
« Reply #15 on: May 14, 2011, 09:10:26 AM »

Quote
The Ham thought for a moment and said, "You know, I love all this stuff. It is perfect. When I want to make a contact I do. I couldn't ask for any more. But, I don't know, it's too easy. Somehow I thought Heaven might be different."

If I lose a contact because of  the vagaries of propagation, insufficiently big antennas or power, or fair competition from other hams, that's fine with me.

If I lose a contact because my SDR crashes just as 7O0OO comes back with "N3OX you're five by ni.." that's pretty much unacceptable.

That kind of failure can happen with any station, SDR or not, and it can happen because an antenna breaks instead of a radio breaking, and all sorts of things.  But that doesn't mean that a buggy radio is "part of the fun," at least not for me.  It's a mistake in station building that should be reduced to very low probabilities.

If you want a quantification on that, I would say a serious radio crash should happen somewhere between one failure per thousand QSOs (casual DXer/ragchewer)  to one failure per hundred thousand  QSOs (serious contester).

An AWKWARD radio is something I'm willing to work around as long as it does the same weird thing every time.  An unpredictable radio is not fun.
Logged

73,
Dan
http://www.n3ox.net

Monkey/silicon cyborg, beeping at rocks since 1995.
W4HIJ
Member

Posts: 367




Ignore
« Reply #16 on: May 14, 2011, 12:04:45 PM »

Hey, I said I wasn't a software guy.  My experience is more on the hardware end of things for sure. I'm not as technically proficient as you guys at understanding this stuff either. But  for some reason, ever since I've been fooling with computers, I've always been very good at putting a bunch of different programs together and making them work for my intended purpose, much like the Flex implementation is done. I don't always understand the why's and how's of it working but I seem to have a knack for "making things play together" Unlike a lot of hams, the prospect of taking my SDR out of the box and hooking it all together, downloading the various programs and configuring them for my needs never daunted me or intimidated me. I quite enjoyed it in fact. I'm a dreamer though I guess, and the dream of being able to turn on my PC and radio hardware and then click one single icon and have a program that comes up with all my Power-SDR functions, logging, satellite/moon tracking, doppler correction, grayline and other propagation tools, digital decoding etc. etc. all ready to go and all communicating with each other within the PC in an efficient manner  doesn't seem so far fetched to me. To me it's one of the great promises of SDR. I mean I don't even think VAC was written with hams or SDR In mind at all, please correct me if I'm wrong. Surely there is a way to "build a better mousetrap". That's all I'm saying.

As for the parable, I agree that things shouldn't be too easy. If it loses the challenge it won't be any fun anymore. BTW that story reminds me of one that took a different tact but in a strange way  is relevant to SDR and what I'm proposing above. I believe is was written by John Troster  W6ISQ ( again if I have the name and call wrong, I apologize and someone correct me) In it the ham had a computer that took on all the mundane task of running and maintaining the station where the OM could concentrate solely on making contacts. Except in the end, the computer took over and started giving the OM commands and task to do while IT operated and worked on IT'S DXCC totals.
Let's be careful here gents!

EDIT: I thought sure the story I remember and referenced above was from W6ISQ but I can't find it in the QST archives under his call. Anyway, whoever wrote it, it was a great story, wish I could find it.
« Last Edit: May 14, 2011, 04:13:29 PM by W4HIJ » Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #17 on: May 15, 2011, 10:13:18 AM »

Whether you like Flex Radio Systems or not, the Flex radios are interesting simply because they are different and we can learn things from both the successes and the mistakes that Flex made in building the radios. I don't think anyone can deny that the transceiver topology developed by Gerald Young and described in his QEX article some years back, i.e. the quadrature switching detector with I/Q output, has had a TREMENDOUS impact on radio design. From that we got the SDR-1000, SoftRock, LP-Pan, and a host of other small, inexpensive radios. People with traditional superhet receivers are using this to add panadaptor and waterfall displays to their radios.

But the problem was to make an SDR that appeals to BOTH the traditional operator AND the bleeding-edge experimenter. The problem is, these desires tend to be in conflict. The first and most obvious problem is that latency/delay is inherent in any SDR design. The laws of physics and mathematics make that the inviolate and inflexible case. The mode most obviously impacted by this is CW. CW, especially full-QSK CW, depends on things happening with sub 10ms delay. It must be well below the threshold of perception for the human brain in order to be perceived as "instantaneous", something needed for QSK. This is clearly an area where an analog radio is going to have an edge, especially if you generate CW on the analog side rather than on the digital side of things. So to build an SDR that meets the desires of the traditional operator is a really hard problem. You also have to make it look and operate enough like a traditional superhet that you don't have such a huge learning curve that you drive [too many] people away. That is a really hard problem too.

OTOH, you have guys like me, the technical hacker, for whom the experimentation with radio technology and radio propagation far outstrips the ability to make a contact. Personally I am not interested in a contact for a contact's sake. I heard OM3TWM on 17m yesterday. I hardly ever hear Europe here. Oh, I called him a couple of times but then went on to something else. It just wasn't that important to me. OTOH, I will sit and watch the doppler shift on the CHU carrier for hours, trying to understand what the ionosphere is doing. I suspect that some of you might consider that crazy. :-)

To someone like me, PowerSDR doesn't go far enough. To me PowerSDR looks WAY too much like a traditional superhet. It is not easily extended to actually make the modem/CODEC an integral part of the receiving system. The whole thing with VAC and virtual serial ports seems to me to be a particularly inelegant hack. OTOH, when I want to get a digital I/Q signal to a new modem, I can do it with PowerSDR and the Flex radios.

But I understand why it is there. It provides necessary backward compatibility so we have a springboard to get people using SDR as they learn what is REALLY possible.

So Flex has gotten themselves caught in the middle. Heard the phrase, "Jack of all trades, master of none?" That is where Flex and PowerSDR sits right now. It is very easy to find fault with pieces of PowerSDR and the Flex radios (as demonstrated by some people here who have made it their pastime to do). For example, I just ranted on the Flex lists about something that has been driving me crazy about PowerSDR from day-one, something that is going to be exacerbated by the new tuning knob.

Regardless, the strength of any SDR, Flex included, is that it can be changed to do something new. When someone comes up with a new idea, the radio can be adapted by changing the software. I know I keep harping on the importance of CW Skimmer but it demonstrates that there are significant new applications that will be made possible by the existence of a wide-bandwidth receiver. (I am not sure that the Flex 5000's 160kHz bandwidth is wide enough but it is a good start.) No one has even thought about the possibility to do wide-bandwidth transmit yet. What if you could hold several simultaneous QSOs? What if you could send different EmComm data to several receivers simultaneously from a single transmitter? SDR makes that possible. Traditional superhet transceivers need not apply.

Personally I think that people need to stop looking at SDR as the future embodiment of the traditional transceiver. If what you want to do is ragchew on CW, it is going to be almost impossibile to beat the current crop of hybrid analog/DSP transceivers. They are fantastic implementations of narrow-band CW rigs. If that is your criterion, then there is no question that the Yaesu FTDX-5000 is better than the Flex or any of the other QSD-derived radios that run PowerSDR. OTOH, if you want to run CW skimmer, you need a wideband receiver with an I/Q output. The only way you are going to get that is with some kind of SDR.

So it really doesn't come down to "this is better," or, "that is better," you have to say, "This is better for this particular mode, and that is better for that other mode."

But one thing is true about SDR -- you can change it. If you discover/develop a new mode, you can change the software to change the behavior of the "radio" to enable that new mode. Radios defined by hardware just can't do that. Certainly you can quibble about the details, e.g. "The FCC might not let us do that," or, "The manufacturer might not release the code so I can make changes," or, "I don't like firewire and an external computer," but that does't change the basic fact that SDRs are mutable and analog hardware radios aren't.

One thing we can do it pressure manufacturers into making their software/firmware public. Yes, that will be a fight with the likes of the FCC and the PTTs (extra-US regulatory agencies, not the transmit button on your mic), but it is what will make your future SDR adaptable to new functions. Frankly, we need to make it clear to the likes of Flex, Elecraft, Ten-Tec, Yaesu, Icom, and Kenwood that open software for their radios is IMPORTANT, that they must make the case to the FCC and the PTTs that general software flexibility is CRUCIAL.

73 de Brian, WB6RQN/J79BPL
Logged
M0HCN
Member

Posts: 473




Ignore
« Reply #18 on: May 27, 2011, 06:04:25 PM »

It has always seemed to me that the major upside of SDR is all the fun things you can do with a wideband IF and potentially all the fun new modes it opens up (OFDM for example becomes possible), the downside (at least for a receiver) is that same wide IF bandwidth makes the requirements on the IF strip rather painful, especially if you want to be able to resolve weak signals close (within the passband) to very strong signals, a hundred Khz IF bandwidth is going to need both a superbly linear IF strip and an ADC that is actually linear across a huge range (and that has a full power bandwidth extending the full width of the IF bandwidth - good luck with that!).

The ADC power bandwidth may in fact be a major limiting factor on many current designs as standard audio sigma delta parts do not have what is needed here (And often don't even detail the limitations in the data sheet), a two tone test at near 0dbFS at ~20K can reveal much that is unpleasent.

As far as transmitters go I see (mostly) major upsides as it makes tricks like cartesian loop feedback nearly trivial and that should really help to clean up the linearity of amateur band transmitters which will in tern make some more advanced modulation modes work.

The delay issue can be largely eliminated by increasing the sample rate and by clever filter design to minimize the temporal length of FIR filters, I don't really see analogue having an advantage here given that something like a SSB Quartz filter will have a group delay of a millisecond or so. Of course on a PC where you basically have to process in blocks of ~64 or more samples you may not be able to get ideal numbers but a DSP dev board can usually work sample by sample. If I can put DSP in the vocal monitor circuit in a studio and have it work for the singer it should not be an issue to make it work for a CW operator.

As to the ability to change things, we can already do that in a hardware radio, we just use a hot stick, scope and vector network analyzer instead of a C compiler....

73's Dan (Who thinks that RX IF bandwidth should be as narrow as the operating mode allows, but has no problem putting DSP behind it).
Logged
KJ4SLP
Member

Posts: 22




Ignore
« Reply #19 on: June 12, 2011, 07:18:15 PM »

What a refreshing thread this is!  I stopped posting to the eham forums altogether a year or so ago because my life has enough stress in it without continually being subjected to bullying, posturing, and snide rudeness.  That seems to have gone away, at least in this thread, and the silence is truly a blessed relief.

I'm a statistician, not an engineer, and for that reason I often don't know just exactly what you folks are talking about.  I am also in a fair way of making a fool of myself every time I open my mouth on technical topics.  So mostly I listen and learn.  I drive my good buddy WB6RQN nuts with my inane questions but he remains patient in the best tradition of the amateur radio service.

I'm a very happy, if sometimes frustrated, Flex 5000 owner, though I am nobody's fanboy, nor ever have been. 

My vision of the future is a piece of extremely versatile hardware for which interested parties can write a whole host of applications.  The joy of the Flex system is its changeability, yet, as WB6RQN and I were saying just this afternoon, already we are coming up against limitations of the hardware.  It seems to me that hardware designers have to find new and better ways to let the software do its thing without "the box" getting in the way.  Am I tilting at windmills here?  Is this the impossible dream?  I've been told (sometimes patiently, sometimes not) that there is just no way (short of rewriting the laws of physics) to build a single box that will do every existing mode--plus a bunch of new ones--and do them everywhere from LF to SHF.  I have to take your word that can't be done but one can dream.

Keep up the conversation, folks.  For once, we have something actually worth reading.
Logged
W6RMK
Member

Posts: 660




Ignore
« Reply #20 on: June 13, 2011, 06:39:20 AM »


Am I tilting at windmills here?  Is this the impossible dream?  I've been told (sometimes patiently, sometimes not) that there is just no way (short of rewriting the laws of physics) to build a single box that will do every existing mode--plus a bunch of new ones--and do them everywhere from LF to SHF.  I have to take your word that can't be done but one can dream.


It's like racing. how fast do you want to go?  It is possible to build a box that will do what you want, but it's pretty expensive.  So, we have boxes that do pieces of it, since that makes the box cheaper.
Logged
KD8K
Member

Posts: 14




Ignore
« Reply #21 on: June 13, 2011, 08:29:37 AM »

What I see in the future is mainstream transceivers migrating more and more towards SDR technologies, with touch screen consoles, and maybe some knobs for basic functions (such as volume, etc), implementing more and more software into their designs. Perhaps similar to current ICOMS, only with touch screens, and fewer knobs. This would enable further integration of software defined technologies in them with features such "Point and Click" tuning via touchscreen and user interface changes via software, all in a self contained unit.

 Perhaps a direct sampling SDR with a touchscreen console and a few knobs in a self contained unit will be seen in the not too distant future.

John
« Last Edit: June 13, 2011, 12:56:44 PM by KD8K » Logged
M0HCN
Member

Posts: 473




Ignore
« Reply #22 on: June 13, 2011, 11:46:55 AM »

That is fine for a rig sold expressly as a SDR,with all the compromises for conventional narrow band working that implies, but it will not compete with a conventional radio for digging weak CW out of a massive pileup on say 160M because the dynamic range will not be there. That few hundred Hz wide first IF filter in a conventional rig is there for a reason. You really need both to be available, SDR for doing what it does and a conventional narrow IF set for when that is appropriate.
Maybe a SDR 'second' receiver so you have the best of both worlds in one box.

Now personally I despise touchscreens for anything that needs frequent use, for all that things like AGC parameters and other occasionally tweaked things could usefully go there, but from an operator ergonomics perspective I need main tuning/rit/volume/IF Gain/band selection/mode and possibly a few other things on hardware controls I can find without looking, all the infrequently used stuff can go to touchscreen.   

Separating the control surface from the radio is however an obvious direction to move in, a simple serial link between the two should be nearly trivial. 

Regards, Dan.
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #23 on: June 13, 2011, 01:56:13 PM »

What I see in the future is mainstream transceivers migrating more and more towards SDR technologies, with touch screen consoles, and maybe some knobs for basic functions (such as volume, etc), implementing more and more software into their designs.

I think this gets into what I consider to be a misconception. It is easy to confuse "software defined" with "software controlled". As soon as you start talking about knobs and touchscreens you are talking about software controlled radios. Now a software controlled radio could be an SDR but as someone who has interfaced something like HRD or flrig to a modern (but not SDR) rig can tell you, it doesn't have to be.

Software defined radios have their RF, IF, and CODEC functions performed in software and can be changed by changes to software. This has nothing at all to do with how you control the radio. In fact, as SDR becomes better defined, I expect there to be fewer controls of any sort beyond possibly audio gain and perhaps frequency/channel. Everything else will be controlled by the CODEC which will redefine the functioning of the RF and IF portions to meet its own needs.

73 de Brian, WB6RQN/J79BPL
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #24 on: June 13, 2011, 02:16:16 PM »

That is fine for a rig sold expressly as a SDR,with all the compromises for conventional narrow band working that implies, but it will not compete with a conventional radio for digging weak CW out of a massive pileup on say 160M because the dynamic range will not be there.

I think you may be mistaken. Now the only SDR with which I have extensive experience is the Flex series. (I also have several Softrock boards but don't really count them as they are just cost-reduced versions of the Flex SDR-1000.) From a pure measurement point of view of testing for dynamic range, the Flex radios perform extremely well. For close-in signals, i.e. for situations where the interfering signal(s) fall within the passband of the roofing filter, they outperform just about any other rig. (I say "just about" because I haven't done any testing but I suspect with about 98% certainty that the Flex 5000 will outperform the current "best" radios in this scenario. You will notice that, unlike the top-of-the-line narrow-band radios, the Flex's DR does not change regardless of spacing between desired and undesired signal.) So it is possible for a wide-band SDR to have good dynamic range.

What I think you are thinking about is what happens when you have many signals at the same time impinging on the ADC. Of course, those signals sum and you will end up with random amplitude peaks that can saturate the ADC. The question is, how often does that happen? How often do you get that level of signal difference between noise floor and a signal or combination of signals that will actually cause clipping of the ADC? Remember, if you have an elevated atmospheric noise floor you can opt for an attenuator to bring the atmospheric noise floor closer to the system noise floor and thus regain all your dynamic range.

But no doubt narrowing the passband prior to the ADC will reduce the amount of unwanted power reaching the ADC and thus allow for greater amplitude differences as you have fewer peaks to sum.

But now here is a question for you: how is this any different from an analog radio? As I learn more and more about DSP it seems to me that there is really no difference between analog and digital (besides digital being more precise and quantized on something higher than a quantum level). Claude Shannon implied that many years ago and that seems to be holding up as well the the laws of Thermodynamics. (There are those who suggest that there is no difference between Information Theory and Thermodynamics -- which in and of itself suggests some very interesting things about our Universe.)

So, I don't think you can separate analog and SDR radios in this respect. I think they both have precisely the same limitations under the same conditions.

73 de Brian, WB6RQN/J79BPL


Logged
KD8K
Member

Posts: 14




Ignore
« Reply #25 on: June 13, 2011, 02:43:18 PM »


I think this gets into what I consider to be a misconception. It is easy to confuse "software defined" with "software controlled". As soon as you start talking about knobs and touchscreens you are talking about software controlled radios.


And I consider it a misconception that an SDR has to run on Windows with an external computer ; ). I was not referring to a software controlled transceiver, but a transceiver that does detection, AGC, Filtering, using Software, instead of hardware, thus being "Software Defined". An SDR can indeed have knobs if the knob digitally controls software, at least in my view Smiley.

I find it interesting that there are projects that add knobs, and touch screen controls to the Flex and yet it is still considered an SDR. What would be the difference if these technologies were combined into a self contained box? Would it no longer be an SDR?
« Last Edit: June 13, 2011, 03:41:54 PM by KD8K » Logged
W4ZV
Member

Posts: 84


WWW

Ignore
« Reply #26 on: June 13, 2011, 04:00:48 PM »

That is fine for a rig sold expressly as a SDR,with all the compromises for conventional narrow band working that implies, but it will not compete with a conventional radio for digging weak CW out of a massive pileup on say 160M because the dynamic range will not be there. That few hundred Hz wide first IF filter in a conventional rig is there for a reason. You really need both to be available, SDR for doing what it does and a conventional narrow IF set for when that is appropriate.
Maybe a SDR 'second' receiver so you have the best of both worlds in one box.

Correct on both counts Dan.  Perfomance (IMD, BDR, phase noise, etc) does matter and there are many alternatives for SDR-based bandscopes.  There's no reason you cannot have the best performance AND a good waterfall (e.g. Skimmer).

A few related comments from ON4UN's latest edition of Low-Band DXing:

##################################################

Top Band Equipment Survey Results  (page 2-31)

The total group of 397 stations listed 493 rigs; for the
top group 112 transceivers were listed by 88 stations. The
most significant difference between the total group (397 stations)
active on 160 meters and the top group (stations having
worked at least 225 DXCC countries) is the fact that the K3
and the Orion are used by a higher percentage of the sample
(see Table 2-4).

Table 2-4  Transceivers Used by 160 Meter Operators

Rig, Total Group, Top Group

Yaesu FT-1000 (D-MP) 26% 28%
Elecraft K3 13% 16%
Icom IC-756 12% 6%
Ten-Tec Orion 9% 11%
All Kenwood equipment 8% 7%
Other Icom equipment 7% 6%
Yaesu FT-2000 6% 6%
Icom IC-7800, IC-7700 5% 6%
Other Ten-Tec equipment 5% 4%
Other Yaesu equipment 3% 4%
Elecraft K2 3% 5%
FlexRadio SDR <1% 0%
Various 1% 1%
Homemade 0% 0%

SDR Hardware  (page 3-13)

FlexRadio Systems (www.flex-radio.com) offers a line
of complete SDR transceivers. In January 2009 I did a survey
among over 2000 active low band DXers (see Chapter 2, Section
20). To my amazement only four out of over 400 stations
(less than 1%) that responded use a FlexRadio transceiver.

##############################################

73,  Bill








 
Logged
WB6RQN
Member

Posts: 484




Ignore
« Reply #27 on: June 13, 2011, 04:41:14 PM »


I think this gets into what I consider to be a misconception. It is easy to confuse "software defined" with "software controlled". As soon as you start talking about knobs and touchscreens you are talking about software controlled radios.


And I consider it a misconception that an SDR has to run on Windows with an external computer ; ).

Who thinks that? Windows and an external computer have nothing to do with SDR. Certainly one could do the processing on a general-purpose PC platform but it isn't necessary and I don't think that I or anyone else suggested that was the case. More to the point, that doesn't have anything to do with what we are discussing.

Quote
I was not referring to a software controlled transceiver, but a transceiver that does detection, AGC, Filtering, using Software, instead of hardware, thus being "Software Defined". An SDR can indeed have knobs if the knob digitally controls software, at least in my view Smiley.

I was just trying to make sure we were talking about the same thing. Many people seem to think that being an SDR has something to do with a computer screen and it does not. The UI really is a separate issue.

Quote
I find it interesting that there are projects that add knobs, and touch screen controls to the Flex and yet it is still considered an SDR. What would be the difference if these technologies were combined into a self contained box? Would it no longer be an SDR?

Of course it would still be an SDR. Being an SDR has nothing to do with how you control it, hence my comment about separating "software controlled" from "software defined".

73 de Brian, WB6RQN/J79BPL
Logged
KD8K
Member

Posts: 14




Ignore
« Reply #28 on: June 13, 2011, 05:42:43 PM »


I think this gets into what I consider to be a misconception. It is easy to confuse "software defined" with "software controlled". As soon as you start talking about knobs and touchscreens you are talking about software controlled radios.


And I consider it a misconception that an SDR has to run on Windows with an external computer ; ).

Who thinks that? Windows and an external computer have nothing to do with SDR. Certainly one could do the processing on a general-purpose PC platform but it isn't necessary and I don't think that I or anyone else suggested that was the case. More to the point, that doesn't have anything to do with what we are discussing.


Perhaps a misunderstanding on my part. I understood your reply to my comment as saying that radios with knobs and touchscreens can only be software "controlled" and not "defined". I apparently did not understand your reply correctly. Anyhow, I think we are now on the same page. My comment was meant to relate to the theme of the thread "SDR and future implementations" in that I see future implementations that will include not only PC based implementations but will also include the larger manufacturers using SDR with touchscreens as the user interface.

73,

John
« Last Edit: June 13, 2011, 07:00:28 PM by KD8K » Logged
M0HCN
Member

Posts: 473




Ignore
« Reply #29 on: June 13, 2011, 08:02:28 PM »

I suspect that we may see a range of SDRness in the market, starting at one end with direct digitization right at the aerial and proceeding all the way to no DSP at all at the other extreme.
I could certainly see  several points in a typical RX where you could (with varying engineering tradeoffs) decide to put the converter.

Question: Does SDR necessarily imply wideband input to the ADC, or is an equally valid implementation to only digitize after cutting the IF bandwidth way down? After the IF amplifier/AGC? Actually at baseband?
Each of these trades off some possible software capabilities for (probably) better BDR.

Now I have to admit to having PA3AKEs awesome receiver board and quartz set on order but I am giving serious thought to going DSP for the detector or possibly doing the DSP at baseband, in any case it will be after some brutal IF filtering, so I probably cannot get much of a waterfall out of it as the IF may have been butchered down to a mere 250Hz wide by the time the DSP gets a look.

Does a DSP based demodulator behind a conventional front end still count as SDR?

Just some thoughts to ponder.



Regards, Dan.
« Last Edit: June 13, 2011, 08:31:16 PM by M6ATV » Logged
Pages: Prev 1 [2] 3 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!