Transmitter Modulators

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Transmitter Modulators

Post by MrAl » Fri Mar 09, 2012 5:21 am

Hi Robert,


Well, if i said i can send a particular kind of message signal content using a bandwidth of 200kHz and i could send that same content using a bandwidth of 100kHz, which one is better.
LEDs vs Bulbs, LEDs are winning.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Transmitter Modulators

Post by Robert Reed » Fri Mar 09, 2012 6:28 pm

MrAl
Assuming no degradation in fidelity, obviously 100 KHz has more appeal due to reduced spectral usage. What are we getting at here?

On a side note:
Do you get the feeling that we are two lonely echoing voices in an empty ware house? This forum is getting creepy with the lack of activity (any where) :sad:

User avatar
CeaSaR
Posts: 1698
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Transmitter Modulators

Post by CeaSaR » Fri Mar 09, 2012 8:20 pm

Robert Reed wrote:"But also i think worth mentioning is that just because the bandwidth of some sort of modulated transmission is low it doesnt necessarily mean some sort of better quality. I can send audio with a 2kHz bandwidth but it will take 10 times longer (or more) than real time to send."

I don't understand this paragraph. Why would a lower bandwidth indicate higher quality to begin with? Also isn't all radio transmission in real time?
Isn't BPSK used more for digital transmission rather than analog? I mean, it is 1's and 0's. And with digital,
MrAl's statement is akin to the difference between the old Dial-up and today's modern super high speed
cable/FIOS/DSL/T1, etc. They all send the same packets, just at much differing speeds. And since we are
talking digital, the fidelity is in the ability of the receiver to convert that data stream back into the original
packets for conversion. That means error correction, redundancy, etal. So, for bandwidth, BPSK can be very
narrow as it only needs to deal with the 2 states - all others outside of the tolerance can be ignored.

Or am I off course again?

CeaSaR

PS - I noticed the lack of activity also. I have been following every day. Wish more people would participate.
I even try to write about stuff in the magazine and the replies from authors are painfully slow at times (like now).
Hey, what do I know?

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Transmitter Modulators

Post by Robert Reed » Fri Mar 09, 2012 10:01 pm

Hello Caesar
Well by golly there is life out there. :grin:

MrAl's statement
"But also i think worth mentioning is that just because the bandwidth of some sort of modulated transmission is low it doesnt necessarily mean some sort of better quality. (indicates the suggestion that low bandwidth it might mean better quality)

My Query
Why would a lower bandwidth indicate better quality to begin with?its usually just the reverse. (Wanted to know how he arrived at this).

Your reply
MrAl's statement is akin to the difference between the old Dial-up and today's modern super high speed
cable/FIOS/DSL/T1, etc.


My response
Exactly! Due to telco lines very low bandwidth capability and cables very high bandwidth capability and wider yet through microwave transmission(air). What I am getting at is there was never any reference to low bandwidth relating to better fidelity.And remember, we are talking about all modes being transmitted through the SAME medium-AIR.

Yes BPSK is only for digital transmission. Any abrupt shift type of transmission, PSK.FSK.ASK and others are strictly digital. Also as to digital being only ones and zeros,They are still pulses that require a lot of frequency components in their makeup.That and 180 degree shifts add up to a lot of bandwidth and high speed data rates kick it wider yet.
How wide?-That is the subject of my post and still a mystery to me.
I am sure you have heard the old phrase when it comes to amplifiers,receivers or any device processing an input signal -" Garbage in; Garbage out". So the sending unit plays a big role in overall system performance.

User avatar
CeaSaR
Posts: 1698
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Transmitter Modulators

Post by CeaSaR » Sat Mar 10, 2012 9:45 am

Ok. I think I am getting to both halves of the conversation between you and MrAl.

You are talking about Fidelity, which can be termed as the ability of the TX signal to remain
as pure as possible through the transmission and ultimately through the receiver and output.
MrAl is actually talking more about efficiency, given the same signal with different throughput
speeds (low bandwidth vs. high bandwidth), allowing acceptable degradation within limits.

Yes, you need a certain amount of Fidelity to recreate the digital signal at the end of the line,
but that Fidelity is, as I believe it, nowhere near what you need for the proper recreation of
analog audio. For analog, lets look at AM vs. FM. Regular AM has a bandwidth of what, 4k-5k Hz?
And that is usually monophonic. Fm runs from 50-15k Hz with a 19k Hz stereo pilot signal. This
signal includes both channels in the original 50-15k Hz "data" band. Now this is a 20k Hz bandwidth
(plus whatever it is that the FCC requires for station separation). So far, all I am talking about is
bandwidth. Fidelity is something different. AM, since it was originally designed for voice, can have
an acceptable distortion of, what, between 5-10% and still be recognizable and decoded by the
human brain. In addition, you don't need the full range of the audio spectrum to do that. That is
why AM is so low-fi. FM needs to be at a much greater fidelity because it was designed for music
reproduction. So now you are at or below a 1% distortion rate with a much greater audio range.
Now you have the need to only reproduce the signal faithfully - not color it with weird swings in
the modulation (not counting compression, bias and all the other things done to improve the signal).
So now we have hi-fidelity FM all within a 20k Hz bandwidth.

Digital, if I can relate it to CD redbook standards (yes, a loose correlation, but I think a pertinent one),
needs to be at twice the data rate plus "a little bit" in order to faithfully recreate the data. By that
"standard", a 300 Baud data rate (yes, Baud. 1985 BBS anyone?) would need a bandwidth of around
700 Baud. Conversely, a 900 Mhz should need over 1.8 GHz, but in the real world we know it doesn't.
Now, this is just bandwidth. Fidelity is the same for both speeds, barring spurious noise. And spurious
noise will affect both systems differently. A 1 mS burst may take out 1 or 2 packets of info on the 300
Baud system whereas it could take out 1000 or so packets on the 900 MHz system. The remaining
packets are all intact and the rest of the transmission is flawless. In this case, the fidelity of the slower
system is better than that of the faster one. The speed sucks, and no one would stand for that anymore.
And to take this particular case one step further, the efficiency number would be better for the slow system
in respect to bandwidth. Same data on a much more narrow channel. Other efficiency criteria, such as
overall system speed, would blow that "advantage" away. This is where your question comes into play.
The designers of the system have to weigh the tradeoffs of bandwidth vs. efficiency vs. reliability etc.
in order to settle on a system.

Hopefully, everyone is as confused as I am now. :shock: I think I have gone over both you and MrAl's points
in an effort to get to a common understanding between all. This layman will now sit back and await the
great discussion that is sure to follow.

Thank you, thank you. I'll be here all week. :wink:
CeaSaR
Hey, what do I know?

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Transmitter Modulators

Post by MrAl » Sat Mar 10, 2012 1:27 pm

Robert Reed wrote:MrAl
Assuming no degradation in fidelity, obviously 100 KHz has more appeal due to reduced spectral usage. What are we getting at here?

On a side note:
Do you get the feeling that we are two lonely echoing voices in an empty ware house? This forum is getting creepy with the lack of activity (any where) :sad:

Hi Robert,

I am still trying to understand what you meant when you said "they are able to achieve a modulation index less than 1".
Then we talked about bandwidth relating to this and it sounded like you implied that lower bandwidth was better, so i thought i would mention that lower bandwidth doesnt always mean better. I could have misunderstood something you said i guess.

As far as radio transmission being in real time, nobody really ever made that statement i dont think because when it comes to modulation techniques it all depends on what we agree on beforehand. If you agree that i will send you information as radio signals that contains audio at 1/2 real time scale (slowed down by a factor of 2) then all you have to do is decode it that way. It's true that in a given time i'll only be able to send half the audio that i would be able to send in real time, but that may not matter for relatively short messages. As long as we both know what is being sent we're good to go.
The Mars robots have to be programmed with instructions several minutes or more beforehand so that they know what to do because there isnt time for the human operators to see what they are seeing and make the necessary changes. Thus quite a bit of prediction has to be included in the programming. For a human to human audio 'real time' transmission, it would take an average of 11 minutes to get back and forth. Thus, if i say "Hello there Robert on Mars", it's going to take 11 minutes for you to hear that, and if you reply, "Hello MrAl on Earth", it's going to take 11 minutes for me to hear that for a total of 22 minutes between the time i send and the time i get your reply. In this case it doesnt matter as much if we use real time or 1/2 real time because real time will take 22 minutes and 1/2 real time will take a tiny bit over 22 minutes (like 22 minutes plus 4 seconds) for the reply.

Yes im still waiting for more people to join the board. Ceasar joined the discussion too so we have three people now :smile:
LEDs vs Bulbs, LEDs are winning.

User avatar
CeaSaR
Posts: 1698
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Transmitter Modulators

Post by CeaSaR » Sat Mar 10, 2012 5:33 pm

Well, I joined in, backed off, rejoined, and am now awaiting more clarity.

CeaSaR
Hey, what do I know?

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Transmitter Modulators

Post by Robert Reed » Sat Mar 10, 2012 7:09 pm

Hi MrAl
"I am still trying to understand what you meant when you said "they are able to achieve a modulation index less than 1".
Then we talked about bandwidth relating to this and it sounded like you implied that lower bandwidth was better, so i thought i would mention that lower bandwidth doesnt always mean better. I could have misunderstood something you said i guess."


First of all, I didn't say this. The "they" refers to a Quote from the original article. Secondly (in regards to bandwidth), I did say this and that it was a good thing- but only in regards to Spectra conservation and not highest quality,which I had again re- mentioned several posts back.Maybe you missed this.If you did,I can understand where the confusion was.

Caesar (like it or not,you are in the group now :grin: )
What I have been trying to demonstrate for a while is the relationship between fidelity and band width and all the variables represented there. Modulation indexs tell a lot of information about that. Fact :
High fidelity (good) requires more bandwidth (bad)
High speed (good) data rates require more bandwidth (bad)
AM is every bit as high quality as any other mode
There are FM radio services that are inferior to AM
The quality for either is set by design for the service required

"Digital, if I can relate it to CD redbook standards (yes, a loose correlation, but I think a pertinent one),
needs to be at twice the data rate plus "a little bit" in order to faithfully recreate the data."


I have never heard of this one before. Are sure it wasn't referring to the Nyquist rate(2.2:1) for sampling without ambiguity?

I get Your's and MrAl's points about the real time issue and have no argument with that, But we are talking about very hi-speed data rates through an AIR medium and see no issue for any snail mail techniques needed since we are not limited in band width such as Line,Cable or any devices that would put an upper limit on that.This is the very reason that the SHF (3000 to 30,000 GHz) band is so popular for this type of transmission is because carriers in that range can support the bandwidth required for the super speeds that data is transmitted today.However that doesn't mean that we can splurge on bandwidth just because it's there. No we still need to reduce it as much as practical due to the over crowded Spectra.

Everybody
Well this has been an interesting post so far, but I think we are getting a little off topic. As far as radio transmission goes, I feel that I am pretty well versed on the subject and didn't want to rehash too much on that. My main point is what the actual spectrum of a transmitted BPSK would look like. This was only to see what the signal would look like and compare it to an analog transmission.Then I could somewhat deduce from what I see to what may be required to send it.
Out of desperation, I spent a little time on the internet this morning trying to research this and as usual came up empty handed. just bits and pieces here and there that never got to the "meat" of things. And Wikipedia was not only horrible but were flat wrong on some topics.

User avatar
CeaSaR
Posts: 1698
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Transmitter Modulators

Post by CeaSaR » Sat Mar 10, 2012 11:34 pm

Robert,

It has been a long time since I read the RedBook CD standards and how they were derived.
Yes, it is Nyquist who presented the 44.1kHz sample rate. From memory: the extra bit of
sample rate was added because they had tried the 40kHz sample rate and had some bad digital
artifacts. Upping the rate to that small bit of oversampling cured those artifacts. I think I may
have been off on that analogy. Not sure why I thought that. :/

Back to the original question. I just did a search with "spectrum analysis of bpsk transmission"
as my criteria (on Yahoo!) and came up with the 3rd one down, Physicsforums.com. The 3rd
post down shows this link, which I hope will help you in your quest. Figures it comes from QSL.net.
Is this sort of what you were looking for?

CeaSaR
Hey, what do I know?

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Transmitter Modulators

Post by Robert Reed » Sun Mar 11, 2012 9:20 pm

I had seen that using a different search engine and it was similar to several other sites that looked hopeful. Unfortunately, they all displayed spectra which looked simulated and with very poor resolution. Also very scanty info on the display - 1KHz carrier? Even if that was a normalized number, it still was inadequate to glean much info from it . But thank you for your efforts Caesar.

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Transmitter Modulators

Post by MrAl » Mon Mar 12, 2012 1:51 am

Hi Robert,

Why dont you try using the LT Spice simulator and do your own frequency analysis? They have menu commands and dot commands that give you Fourier and FFT.
If you've never used it before i can help you get up and running and with the waveform generation necessary to simulate a transmission.
The FFT shows a couple significant harmonics and the rest tend off to very low levels, getting lower and lower as the frequency increases.

Im still not entirely sure what you are after here. You've asked for the bandwidth and i gave you two things to think about but you're still asking. This means we dont completely understand each other, so perhaps you would get more info by doing a couple of simulations yourself, and at the same time have some fun with the simulator :smile:
LEDs vs Bulbs, LEDs are winning.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Transmitter Modulators

Post by Robert Reed » Mon Mar 12, 2012 9:15 pm

Hi MrAl
Thanks for the offer of help with the simulator, but unfortunately I neither have these programs nor knowledge of their use. Call me old school if you wish, but I just prefer working my designs from scratch and then prototyping for proof of performance. And since I do a lot of work through VHF and into UHF, the circuit board layout become key issues. Also, in conversations I have had in the past with people that are far more knowledgeable than me claim that simulators will not give me clear cut information on high frequency complex modulation methods. Just not like a high quality spectrum analyzer display, especially when you can easily adjust a lot of display variables.
I think whats happened to our postings here is that in trying to analyze the situation, we probably got side tracked on a lot of ancillary issues and started to dwell on them. Not that those issues were not important and were interesting discussions but they do not bring me any closer to a final answer.I have run enough spectrum analyzer displays on modulated carriers (some fairly complex) to feel comfortable in making an analysis of them, hence my request for an actual BPSK display.
I think we may be at a dead end here due to a lack of specific information on the web. This surprised me as just about anything is available. I'll bet its hiding somewhere maybe under different search titles. If myself or you or Caesar come up with something in the near future,hopefully we can open this post again and share the info.

PS- again I am shocked by the lack of activity on this forum.I think this has been the only active topic for the last 10 days and thats for all categories!

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Transmitter Modulators

Post by MrAl » Tue Mar 13, 2012 1:22 am

Hi Robert,

Yeah low activity here. Strange really.

So are you saying you dont accept the Professor's lecture where he says that the bandwidth is equal to the symbol rate?
LEDs vs Bulbs, LEDs are winning.

gerty
Posts: 290
Joined: Mon Dec 06, 2004 1:01 am
Location: Tennessee
Contact:

Re: Transmitter Modulators

Post by gerty » Tue Mar 13, 2012 6:44 am

[quote]Yeah low activity here. Strange really.

[/quote]
I would participate if I was knowledgable in the subject matter, but in cases like this I find it better to keep my mouth shut :grin:

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Transmitter Modulators

Post by MrAl » Tue Mar 13, 2012 8:59 pm

Hi gerty,

That's what i should have done too :smile:
Admittedly this isnt my strong area. Forgotten too much i guess.
I know what the harmonics are, but there's something else going on with the technology because apparently they dont think that all of the harmonics have to be considered, maybe because they fade off so fast. The harmonics go to infinity, but we have to face it that if we have a fundamental that is even as low as 0.5v peak if we have a harmonic that is 0.001v peak we may not want to consider it, and another harmonic that is 0.000001v peak (1uv peak) we probably have no real application where this will matter. The true harmonic content gets lower and lower, say 0.001, 0.0001, 0.00001, 0.000001, etc., as frequency increases, so at some point we wont care anymore.
In older power converters they usually didnt go above the 7th harmonic when they designed them, simply not considering anything beyond that. It's not that it is not there, it's just that it doesnt have as much influence over the finished product as the lower ones do.

It's just like anything else...
If someone says, "I get 25 miles to the gallon with my new car", and then someone else stands up and yells, "No, you get 25.000001 miles to the gallon with that new car". At some point we dont care anymore because it really really doesnt matter in any practical way.
LEDs vs Bulbs, LEDs are winning.

Post Reply

Who is online

Users browsing this forum: No registered users and 12 guests