1. Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
    Dismiss Notice

DSB-SC versus SSB-SC demodulation

Discussion in 'Radio and Communications' started by Kerim, Apr 17, 2016.

  1. Kerim

    Kerim Member

    Joined:
    Nov 2, 2011
    Messages:
    65
    Likes:
    6
    Location:
    Aleppo city - Syria
    When I was at the university about 37 years ago (end of 70’s) I used reading/hearing:
    Recovering the modulating signal of a double sideband suppressed carrier signal is much harder than of a SSB-SC one.

    But on the other hand, it sounded rather illogical, to me in the least, that receiving two sidebands (symmetrical) could be worse than receiving just one of them!

    So, as an MS thesis, I designed and built (after many failures during about 5 months) a simple low-cost linear demodulator for DSB-SC.
    Its mains components were:

    1 LM339 comparator IC
    1 CD4046 a PLL IC
    1 CD4013 as flip-flop
    1 CD4066 CMOS switches

    Obviously, it works for any AM index; from m=0 (carrier with no modulation) to infinity (DSB-SC) while it has a frequency lock range (a PLL characteristic), as the AFC in FM receivers.
    Also, being a synchronous AM demodulator, it is suitable for recovering hi-fi audio signals as well (as long the receiver bandwidth is wide enough).
    And, if necessary, its components could be integrated in one low cost IC.

    But, due to a financial issue at that time, I had to return home before submitting my work (I just tested its circuit at the laboratory and later I used its topology at home in some of my private projects).

    Now, I wonder if, even in these days, the graduates in communications (around the world) are made to believe that a simple linear AM demodulator for any AM modulation index (unlike the known complex ones, as Costas Loop or using selective LC filters) doesn’t exist (not known yet).

    Edited:
    Sorry, being new here, I noticed now that I had better posting this thread in:
    Forums > Electronics Categories > Radio and Communications
     
    Last edited: Apr 19, 2016
  2. Nigel Goodwin

    Nigel Goodwin Super Moderator Most Helpful Member

    Joined:
    Nov 17, 2003
    Messages:
    39,294
    Likes:
    647
    Location:
    Derbyshire, UK
    I've no idea where you got those ideas from?, but old analogue TV was SSB-SC and its absolutely trivial to receive, no different to AM - the SSB-SC was used in order to squeeze more channels in the available space.
     
  3. Kerim

    Kerim Member

    Joined:
    Nov 2, 2011
    Messages:
    65
    Likes:
    6
    Location:
    Aleppo city - Syria
    You are right Nigel. Reducing the bandwidth is very important in RF communications.

    But, based on your knowledge, which is easier to demodulate (and to generate)... a SSB-SC or DSB-SC signal?
    Thank you.
     
    Last edited: Apr 18, 2016
  4. dave

    Dave New Member

    Joined:
    Jan 12, 1997
    Messages:
    -
    Likes:
    0


     
  5. Kerim

    Kerim Member

    Joined:
    Nov 2, 2011
    Messages:
    65
    Likes:
    6
    Location:
    Aleppo city - Syria

    I didn't give details about my work (above) because there is a possibility that its circuit topology is already known at today's universities though I couldn't find it on the internet.

    On the other hand, I heard that FCC is no more interested in the DSB-SC system for general applications (as for RF amateurs and in the aviation field), likely because they believe (at FCC) such a system needs a rather complex reliable demodulator, besides requiring twice the SSB bandwidth.
    (But, at the same time, the advantage of using DSB-SC instead of SSB-SC is the ease of generating its signal. And if its reliable low-cost linear demodulator is used, the recovery of the modulating signal is also better than in case of SSB for being synchronous and having a frequency lock range of the suppressed carrier).

    For those who may be interested, I posted a brief of its topology on:
    https://en.wikipedia.org/wiki/Talk:Double-sideband_suppressed-carrier_transmission

    For instance, I simulated it (actually updated versions of it) using LTspice (a free linear/digital circuit simulator from Linear Technology).
    On request, it will be my pleasure discussing it and uploading its LTspice files for more analyses.

    Kerim (67 years old)
     
    Last edited: Apr 18, 2016
  6. Nigel Goodwin

    Nigel Goodwin Super Moderator Most Helpful Member

    Joined:
    Nov 17, 2003
    Messages:
    39,294
    Likes:
    647
    Location:
    Derbyshire, UK
    No different - just a simple AM detector.

    I'm presuming by 'SC' you mean the carrier is reduced (which is how TV signals worked), and not removed completely.

    If it's removed completely then it's the same as SSB and you need to reinsert a carrier to demodulate it, or use other of the various SSB methods.
     
  7. Kerim

    Kerim Member

    Joined:
    Nov 2, 2011
    Messages:
    65
    Likes:
    6
    Location:
    Aleppo city - Syria
    I mean by SC... the carrier is indeed removed completely. So, the power of the transmitted signal should be zero during no speech periods. Otherwise, what I am talking about would be, as you said, another simple detector for AM (with carrier as in MW band or reduced).

    Yes, in case of SSB-SC, the suppressed carrier should be generated locally and inserted to the received signal (in a way or another) in order to demodulate it. After all, the single sideband (of SSC-SC) has no information (at all) about the frequency and phase of the suppressed carrier.

    But in case of DSB-SC, the suppressed carrier could be retrieved (frequency and phase) from the symmetry of the received two sidebands. So, there were many different attempts to recover the carrier from the two sidebands but the known reliable solutions (as the linear Costas loop, and now made digital) happen being harder to implement than all other AM detectors (including for SSB-SC). For this reason and to ease the detection of DSB-SC, one of the two sides is filtered out in some receivers, so that the remaining one could be treated as SSB-SC.

    It seems, for the time being in the least, I am the only one who knows that the universal statement (adopted in all universities) "Recovering the modulating signal of a DSB-SC is much harder than of a SSB-SC one"... is false ;) But I will be glad hearing someone (preferably a professor or student in communications) who agrees with me or is interested in this subject (as a theory that can be implemented). I already used, when I was younger, its advantages to build simple scrambled short-range RF links on MW (by varying the frequency of the transmitted suppressed carrier) and FM (by removing the 19 KHz pilot).
     
  8. BobW

    BobW Active Member

    Joined:
    Apr 28, 2010
    Messages:
    516
    Likes:
    54
    Location:
    Canada
    The main difference between demodulating SSB-SC and DSB-SC (at least for audio signals) is that for SSB the phase of the re-inserted carrier doesn't matter and the frequency can even be off by a few Hz. For DSB-SC the re-inserted carrier must be exactly on frequency and in phase or else the recovered signal will be distorted.
     
  9. Kerim

    Kerim Member

    Joined:
    Nov 2, 2011
    Messages:
    65
    Likes:
    6
    Location:
    Aleppo city - Syria
    You are totally right, Bob.
    In fact, the job of the PLL (just one, not two as in Costas Loop) in my simple AM demodulator is to help recovering 'exactly' both the frequency and phase of the carrier from the two symmetrical sidebands; even if it is completely suppressed as it is the case of DSBSC.
    This is based on a little trick that I discovered by a mere coincidence (after about 5 months of failures at the university laboratory) much like what happened to Archimedes when he said "I found it... I found it" ;)
    But it seems it is no more important, in these days, to end, for good, the universal statement (I used hearing of since I was at the university in the 70's) that says:
    "Recovering the modulating signal of a DSBSC signal is much harder than of a SSBSC one".
    To me in the least, it is easier and even more practical due to the additional features that are provided by PLL (as pointed on OP).
    Of course, there is nothing for free; the bandwidth of DSB is twice of the SSB.
     
  10. BobW

    BobW Active Member

    Joined:
    Apr 28, 2010
    Messages:
    516
    Likes:
    54
    Location:
    Canada
    One method that I'm aware of to recover the original carrier is to square the signal. This results in a signal at double the original carrier frequency that can then be divided by 2 to get back the original carrier.
     
  11. Kerim

    Kerim Member

    Joined:
    Nov 2, 2011
    Messages:
    65
    Likes:
    6
    Location:
    Aleppo city - Syria
    Yes Bob, the two methods that are already known since I was young are the Squaring Loop (you mentioned) and Costas Loop (by using two PLLs in quadrature). But, as you know, implementing them is not easy in comparison to demodulation of SSBSC.
    So I am surprised how no one got the idea yet of the simple topology I found since about 37 years ago!
    This explains why you couldn't find on the internet a method simpler than of the Squaring Loop (and of Costas Loop).

    For instance if you are in the field of RF communications, you certainly get the basic idea (trick) of the new method I am talking about here by reading my post on:
    https://en.wikipedia.org/wiki/Talk:Double-sideband_suppressed-carrier_transmission
     
  12. Kerim

    Kerim Member

    Joined:
    Nov 2, 2011
    Messages:
    65
    Likes:
    6
    Location:
    Aleppo city - Syria
    I started this thread 2 months ago to prove (at least indirectly) that, even in science (hence not only in Politics and Religions) most people on earth are made, in a way or another, ready to believe only what their Elite approve/confirm (but these VIPs are likely different for one person to another). They believe them even more than they believe themselves (their personal logic and experiences).

    Therefore I am afraid that I have to remind you that as long all your Elite ;) anywhere in the world have no idea of what I am talking here, what I presented should look nonsense (and I won't be surprised if it sounds, for some, like another conspiracy theory ;) ).

    I personally demodulated, using a simple reliable PLL circuit, the added DSB-SC signal in the stereo FM broadcasting without the need of the pilot (19 KHz which is the half of 38 KHz; the frequency of the suppressed carrier). In fact, about 30 years ago, I made a private phone RF link (on FM band) and modulated the voice signal on 32 KHz (of the popular mini-crystal) as DSB-SC. Obviously, I let the baseband (mono) of the RF FM signal be empty (silent). I also didn't need adding the pilot 16 KHz. So, in a way, I did what we may call a private scrambled phone link (since no receiver was able, at that time and perhaps now, recovering the transmitted voice).

    Finally, we may say; since no one in IEEE, for example, has an idea of the topology of my versatile AM demodulator, all undergraduates in communications (at EE faculties) around the world have to be told that demodulating SSB-SB is much easier than demodulating DSB-SC... though in reality the inverse is true. But who cares ;)

    Have a nice day.

    Kerim
     
  13. Mark Sanders

    Mark Sanders New Member

    Joined:
    Jun 28, 2016
    Messages:
    1
    Likes:
    0
    Hello Kerim,
    Would be intersted in your single PLL circuit used to demodulate DSB-SC.
    For either DSB or SSB demodulation the carrier reinsertion determines the quality of the modulating signal and
    the processes are usually described mathematically. This is a bit of a pity because electronics should be fun and the transmission and recovery of signals is fairly easy to understand when waveforms are drawn and frequency positions shown ! Anyway this reply may be late in time ?
    Mark
     
  14. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    709
    Location:
    Somerset UK GMT
    Kerim,

    I can understand your frustration that no one seems to recognize your invention. I make no comment about its viability or benefits but would just like to make some general comments about your position and reactions.

    In general, if you suggest something new:
    (1) No one will pay any intention to it
    (2) People will say it won't work, without justification
    (3) People will say it is a crazy idea
    (4) People will say that you are crazy
    (5) People will say it is no use because: it is against god's wishes, it will have a negative impact on global warming, it will not be able to make french fries.

    This is a fact of life (which is baffling), but fact it is. It is not due to a conspiracy or attacks on you personally, it happens because of human nature and dogma. But at least, these days, you are unlikely to be incarcerated, tortured, or burned to death for your ideas, in the developed countries anyway. If you read the history of most inventions you will see this clearly illustrated. I am too lazy to list examples of this but may do later when I wake up fully.

    I hate to say this, but you inserted your idea in Wikipedia, where I do a lot of writing and know the main rules. One of the most fundamental and important rules is that the Wikipedia content shall be factual and unbiased, specifically that original research is banned. So, to be blunt, your entry on Wikipedia was inappropriate in the extreme and was bound to be removed from the article main page (hope I didn't remove it):wideyed:. The same thing happens to many other people's entrys. All the same, your comments were left on the talk page. The deletion caused you some concern and now you are concerned about your reception on this thread; don't you notice a common factor here?

    The other thing to understand is beliveability, not just facts. Take an example. Say you won £20M on the lottery but chose to maintain your current life-style, with just perhaps a few luxury items. If you met someone and told them that you were a millionaire- they would not believe you.

    May I suggest that you draw up a schematic of your circuit and post it here along with a full description so that we can all see your creation.

    spec
     
    Last edited: Jun 28, 2016
  15. Kerim

    Kerim Member

    Joined:
    Nov 2, 2011
    Messages:
    65
    Likes:
    6
    Location:
    Aleppo city - Syria
    Hello Mark,

    I wasn't here since many weeks (I don't remember exactly). I thought I was not welcome here. Only today, I received a notification of your post.

    Since you are personally interested in this subject, it will be my pleasure to share my work with you.
    Sorry, but may I know in which field(s) your scientific background is?
    Are you an undergraduate or perhaps a professor in communications?

    Anyway, as you will see from the circuit I made for simulation on LTspice, the solution has just one simple added trick while the remaining PLL parts could be designed as always.

    I like to add that it happened to me as it happened to Archimedes when he shouted in the bathroom "I found it!... I found it!". I noticed this simple trick that I made unintentionally (due to the different fall and rise delays of a comparator) after trying, for about 5 months, many failed designs (topologies) at the lab. At that time, I was insisting that having two (symmetrical) sidebands should lead us to a simpler demodulator than having just one... based on human logic ;) And it was out of question, for me, to end up with a failure because it was I who suggested this project as an MS thesis when I was an undergraduate at the American University of Beirut. But I couldn't submit it on a document (and get the degree) for financial reason, so I preferred to take care, as soon as possible, of my private small business, in electronics; after all, a boss doesn't need a degree if he can have the satisfaction of his customers ;)

    Kerim
     
  16. Chris-239

    Chris-239 New Member

    Joined:
    Sep 2, 2017
    Messages:
    1
    Likes:
    1
    Hello Kerim,
    I would very much like to hear more about your DSB-SC demodulator. To be able to see a circuit would be nice. I am presently working on a standard DSB reduced carrier detector using a pll but your circuit sounds far more interesting. Please contact me to discuss principles of operation.

    Best regards,
    Chris.
     
    • Like Like x 1
  17. Kerim

    Kerim Member

    Joined:
    Nov 2, 2011
    Messages:
    65
    Likes:
    6
    Location:
    Aleppo city - Syria
  18. btgwynn

    btgwynn New Member

    Joined:
    Nov 7, 2017
    Messages:
    1
    Likes:
    0
    Kerim,

    I saw your post on Stack Exchange and tried to reply, but the admins blocked me there.

    I actually worried that you'd been a casualty of the fighting in Aleppo.

    My interest in you work is because I use a modification of Costa's loop in my research, and the use of the clipper in your schematic helped me in my design, and I would give you credit if possible.

    My question is, since you are using double frequency, what makes your circuit functionally different from Costas Loop or Squaring Loop? The product of the quadrature and in line VCO outputs of a Costa's loop makes a double frequency sinusoid, same as in a squaring loop, as I understand. Your circuit allows a simpler implementation by bypassing circuit architecture needed to get the double frequency sinusoid in other methods, as I understand it.

    Hope to hear from you.
     
  19. Kerim

    Kerim Member

    Joined:
    Nov 2, 2011
    Messages:
    65
    Likes:
    6
    Location:
    Aleppo city - Syria
    As you said, it is simpler.

    One PLL only is needed instead of using two with a 90 deg phase shifter (Costas Loop).
    Also it doesn't need any sort of selective filter, hence no coil (LC tank) in case of high frequencies (Squaring Loop).

    So even in early 80's, I was able to design and build short range RF links (for voice) between home and store, based on DSB-SC.

    I did it for three reasons.
    First, being rather poor at that time, I couldn't get a phone line at home. I had one at store only.
    Second, no other receiver was available to demodulate the transmitted DSB-SC signal.
    Third, I was able, in case of MW band, scrambling my RF signal by deliberately varying the frequency of its suppressed carrier (about +/- 5%) at a rate of about 6 Hz (sinusoidal). In case of FM band, I modulated the voice signal on 32.768 KHz suppressed carrier (much like stereo FM on 38 KHz) first but ‘without adding a pilot’. Naturally there was also no receiver to recover my signal. Also my two FM channels (two-way link) sound empty on conventional FM receivers due to the lack of an audio signal that mono FM channels broadcast.

    By the way, using this method, a low-cost reliable integrated circuit (IC) could be made to demodulate DSB-SC. Lately as a hobby, I updated my first design to have a better response and noise immunity and checking it by using LTspiceIV.

    So, it is sad that no one in the world was interested about it mainly when it was most needed 35 years ago.

    As you know, the main advantages of a DSB-SC system are:

    [1] Its modulating circuit is relatively simple.
    [2] It frees automatically its channel band when no signal is transmitted.
    [3] Its average power efficiency is the highest.

    It is very good, therefore, for aviation communications; mainly between airport and nearby airplanes working on the same (suppressed) carrier frequency.

    Unfortunately, FCC had/has no idea about my work (which is natural since I was born and live in a small country). I also heard that FCC, in a way, has banned lately the use of DSC-SC systems for not being practical for general use and hobbyists while it is/was very practical on AM and FM in Aleppo city ;) But, truth be said, its main disadvantage is that it takes twice the band of SSB channel.
     

Share This Page