AM + FM = ?

[Acme]

The last time I studied radio electronics we were still using tubes (valves for you Brits), though the ideas of amplitude modulation and frequency modulation remain the same. Moving to the crux of the matter, is it possible to transmit a radio signal that is simultaneously amplitude and frequency modulated?

[fiveworlds]

Yes

[Acme]

 

Moving to the crux of the matter, is it possible to transmit a radio signal that is simultaneously amplitude and frequency modulated?

Yes

 

Can you expand on that? Perhaps point me to some references or talk about what you know of it? What applications does it have? What advantages/disadvantages over just AM or just FM? I searched for almost an hour using different terms and got nada. Little help?

[fiveworlds]

Perhaps point me to some references or talk about what you know of it?

 

I happen to have a very old book owned by my father on electronics and radio published in 1969 by Nelkon and Humphreys.

 

What advantages/disadvantages over just AM or just FM?

 

None I don't think its used.

In amplitude modulation the Amplitude of the carrier waves is modulated by the audio frequency.

In FM the frequency of the carrier wave is modulated.

These changes in either Amplitude or Frequency of the carrier waveform are then detected by your radio and converted into sound.

[Acme]

I happen to have a very old book owned by my father on electronics and radio published in 1969 by Nelkon and Humphreys.

So can you quote the material from it that deals with combining AM and FM on/in one transmitted signal?

 

None I don't think its used.

In amplitude modulation the Amplitude of the carrier waves is modulated by the audio frequency.

In FM the frequency of the carrier wave is modulated.

These changes in either Amplitude or Frequency of the carrier waveform are then detected by your radio and converted into sound.

So you don't know if its used; gotcha. I know what AM and FM are which is what prompted my question about combining the two methods of modulation in one transmitted signal.

 

Anyone else?

[fiveworlds]

So you don't know if its used; gotcha. I know what AM and FM are which is what prompted my question about combining the two methods of modulation in one transmitted signal.

 

Well I don't see a reason for it being used you radio detects different things to tune into both methods of carrier wave creation. I mean the only reason would be if you wanted to encrypt a signal or make it really hard to detect. Below is an example of an AM wave the radio detects a particular frequency. The radio would need to constantly be tuned in to detect the radio wave as you send it. But you could for instance make a radio that would increment/decrement the frequency it listens to based on a timetable built into the radio.

 

Edited May 2, 2015 by fiveworlds

[Acme]

Well I don't see a reason for it being used ...

I get that.

 

...you radio detects different things to tune into both methods of carrier wave creation.

If you mean that both the transmitter and receiver would be unique for handling a dual-modulated wave, then I agree.

 

I mean the only reason would be if you wanted to encrypt a signal or make it really hard to detect.

Not necessarily the only reason, but I agree that is a valid example of a reason. Then too maybe we would modulate both the frequency and amplitude with the same signal, possibly for some gain in response or other quality desirable in the modulating information.

 

 

 

Below is an example of an AM wave the radio detects a particular frequency.

 

 

None of those look like a combination of FM & AM.

The frequency is constant in all 3 diagrams so they are all amplitude modulated signals.

[StringJunky]

In principle, you could have the same signal content going on the amplitude and frequency parts. AM is good for distance and FM is good for lower noise and quality, so, you could have a more seamless transmission/reception in a wider variety of terrains, environments and distances.

[fiveworlds]

None of those look like a combination of FM & AM.

The frequency is constant in all 3 diagrams so they are all amplitude modulated signals.

 

You are correct. There is one difference in that it has been noticed that if 2 fm waves are transmitting on similar amplitudes the wave with the higher frequency wins.

[Endy0816]

Found this on arXiv:

Combined frequency-amplitudenonlinear modulation: theory and applications

[Acme]

Found this on arXiv:

 

Combined frequency-amplitudenonlinear modulation: theory and applications

Excellent. Danke. Have saved the PDF and will be reading it. Be back with any questions or observations if or when I find them.

 

Addendum:

So the authors did not carry out any experiments and they only point to spin-transfer [nano]oscillator experiments by others as working examples of their combined modulation calculations. (They call it NFAM: Nonlinear Frequency & Amplitude Modulation) Since these micro devices are working in the microwave frequency range it seems reasonable that one could scale-up to a longer range broadcast transmitter. (Oui/no? )

 

It also appears that the authors are looking at the case of a single input signal so we still have the question of whether or not we can have 2 input signals where one is doing the amplitude modulation and the other the frequency.

[Externet]

Acme :

To use an analogy to explain AM and FM; think of a discotheque playing music and its lights varying its luminosity by the audio. They become brighter in louder passages/notes. That is amplitude modulation.

 

Now imagine the lights at a fixed/steady brightness and varying its color at the audio, -Say yellow at silence- Bluer at higher notes, redder at lower notes of the audio. That is frequency modulation.

 

Simoultaneity seems possible, variation in brightness and color; and feeding two different audio programs seems possible; if both detection systems are implemented at the receiver to separate them.

There can be some interference as AM can perform what is called 'slope' detection. The circuitry would have to be designed accordingly to avoid 'crosstalk' which is one signal appearing in the other channel causing distortion, and the FM discriminator with a threshold capable of keeping the FM signal captured at all times, probably with AM well under 100% modulation.

[i have not read Endy's link yet]

[Acme]

Acme :

To use an analogy to explain AM and FM; think of a discotheque playing music and its lights varying its luminosity by the audio. They become brighter in louder passages/notes. That is amplitude modulation.

 

Now imagine the lights at a fixed/steady brightness and varying its color at the audio, -Say yellow at silence- Bluer at higher notes, redder at lower notes of the audio. That is frequency modulation.

Thnx, but as I said, I studied radio and electronics in school as a boy so I understand the principles. It's that understanding of principles that prompted my question about a hybrid modulation. (I understand there is also phase modulation but I'm not familiar with it nor sure if it is a factor here.)

 

Simoultaneity seems possible, variation in brightness and color; and feeding two different audio programs seems possible; if both detection systems are implemented at the receiver to separate them.

There can be some interference as AM can perform what is called 'slope' detection. The circuitry would have to be designed accordingly to avoid 'crosstalk' which is one signal appearing in the other channel causing distortion, and the FM discriminator with a threshold capable of keeping the FM signal captured at all times, probably with AM well under 100% modulation.

[i have not read Endy's link yet]

That's in line with the kind of answers I'm looking for. I gather you are not aware of this actually being done?

I have to wonder if one or another frequency range would be better adapted to a hybrid modulation. And would such a hybrid radio modulate each mode one-at-a time or in parallel? And if one-at-a-time, would you modulate the amplitude first and then the frequency, or the frequency first and then the amplitude?

[EdEarl]

Television modulates audio as FM ad video as AM. See: Wikipedia.

 

IDK for a fact, but I suspect that some military radars use a combination of AM pulse modulation (for target range and direction data) and detect Doppler frequency change of returned signals (for target speed and acceleration). To avoid detection, it might also FM modulate the carrier into spread spectrum.

Edited May 3, 2015 by EdEarl

[Externet]

Acme :

To attempt the hybrid, I would modulate FM first. At the final RF amplification stage, the voltage feeding the FM final amplifier can be varied in amplitude by the loudness of the 'second' channel. Could be a simple transformer with its secondary in series with the DC supply, its primary the 'second' (AM) channel.

 

Broadcast FM radio in stereo, can itself put two independent audio signals on the air, plus another channel called subcarrier, plus the 19KHz pilot. Adding AM would make 4 channels in a single radio frequency transmission.

 

Getting into digital, time slicing of the modulation merges several channels by multiplexing. One slice has data/audio from one channel, and the next slice from another, and... keep counting. Like cellular telephony, where multiple users voices are all in the same radio frequency.

 

Am not aware of the hybrid being done/used as digital took over for multichannel transmissions. One single optic fibre carries multiple channels of television, bidirectional internet, voice services, data, telemetry...

[Acme]

Television modulates audio as FM ad video as AM. See: Wikipedia.

Gotcha. However, that uses 2 different signals each with a different type of modulation and I am asking about 1 signal with 2 different types of modulation. I'm not sure your Wiki's "...the audio and video are combined at the transmitter before being presented to the antenna, but separate aural and visual antennas can be used" is the same thing as I am asking about.

 

IDK for a fact, but I suspect that some military radars use a combination of AM pulse modulation and detect Doppler frequency change of returned signals. To avoid detection, it might also FM modulate the carrier into spread spectrum.

I ran across some reference to sonar using a hybrid modulation earlier but I'm not sure if I saved it.

Acme :

To attempt the hybrid, I would modulate FM first. At the final RF amplification stage, the voltage feeding the FM final amplifier can be varied in amplitude by the loudness of the 'second' channel. Could be a simple transformer with its secondary in series with the DC supply, its primary the 'second' (AM) channel.

Very interesting.

 

Broadcast FM radio in stereo, can itself put two independent audio signals on the air, plus another channel called subcarrier, plus the 19KHz pilot. Adding AM would make 4 channels in a single radio frequency transmission.

Double the fun!

 

Getting into digital, time slicing of the modulation merges several channels by multiplexing. One slice has data/audio from one channel, and the next slice from another, and... keep counting. Like cellular telephony, where multiple users voices are all in the same radio frequency.

 

Am not aware of the hybrid being done/used as digital took over for multichannel transmissions. One single optic fibre carries multiple channels of television, bidirectional internet, voice services, data, telemetry...

My searching turned up a lot of the digital signal processing themes as you mention. I am unsure how they may or may not apply to the analogue scheme I am asking about.

 

Yeah; surrender gracefully the things of youth. Keepin' my ears on.

[studiot]

I think modern military frequency hopping radio uses an entirely different form of modulation, but if they were to use AM for the signal, frequency hopping could count as a form of combined modulation.

 

The combination maths works like this (simplified to show the principle)

 

Consider a carrier wave

 

y = A cos (w_ct + B)

 

y is the instantaneous total signal, which is a function of time.

A = amplitude

Where w_c = angular frequency = 2pif_c of carrier

t = time

 

 

In a simple unmodulated wave, A and B are constants

In AM we make A variable without affecting B and

In FM we make B variable without affecting A

 

 

So if we introduce a second signal frequency w_s = 2pif_s as a second variable Ccos (w_st) and C is another constant.

Then we can substitute this for either A or B to obtain an AM or FM signal respectively.

 

AM is then {Ccos (w_st)} * {cos (w_ct + B)}

 

FM then A cos { w_ct + Ccos (w_st) }

 

A combined AM~FM signal would be {Ccos (w_st)} * { cos[w_ct + Ccos (w_st)] }

 

Now substituting either A or B is enough to convey all the carried information so why would you want to complicate matters by substituting both?

 

Why would you be wanting to modulate with two separate modulating signals?

There are other, easier ways to do this.

One example of the TV signal has already been given.

Another example would be the radio code data which uses a sideband.

 

 

Further the limitations on the amplitude and frequency of the substituting are different and a double substitution would have to satisfy the limitations of both simultaneously.

Edited May 3, 2015 by studiot

[John Cuthber]

What frequency do you tune the am part of the receiver to?

Also, since it's possible to receive an fm signal with an am receiver, how are you going to disentangle the two signals?

 

 

http://www.daenotes.com/electronics/communication-system/fm-slope-detector

[studiot]

Of course there is another alternative.

 

For the frequency modulating signal to itself be AM modulated.

 

This is, of course, what happens in practice insofar as an audio signal continuously changes in both frequency and amplitude.

[swansont]

What frequency do you tune the am part of the receiver to?

Also, since it's possible to receive an fm signal with an am receiver, how are you going to disentangle the two signals?

 

I imagine you could split the signal out of the antenna and send half to your AM receiver and half to the FM.

[John Cuthber]

Yes, you could.

And that signal would be essentially a voltage changing with time. It would be centred on the carrier frequency and it would have side bands.

Now imagine that I switch off the AM part of the transmission, I get a signal that's centred on the carrier frequency and it would have side bands.

OK so, instead I only switch on the AM transmitter and I get a signal that's centred on the carrier frequency and it would have side bands.

 

I'm not sure that the two receivers can unambiguously distinguish between the two sorts of signals.

By way of a "hand-waving argument" in support of this, i invite you to consider the technology of synthesisers (as in electronic music)

The early ones used am synthesis and could, in principle, mimic any sound.

The newer ones use FM and can in principle, mimic any sound.

 

I think the receiver would be in the same position as the human ear listening to 80s pop.

It wouldn't know (and couldn't, even in principle know) whether the signal it received was AM or FM.

 

As far as I can see, you have run out of variables.

[Enthalpy]

Combined amplitude and frequency modulation is used all the time. It's called a constellation, and serves to transmit more bits per second than the available hertz band, when the signal is much stronger than the noise.

 

More precisely, a constellation modulates simulatenously the amplitude and the phase to transmit one symbol chosen among many, where "many" is a combination of one phase among many and one amplitude among many, and the binary log of "many" tells how any bits of information are coded by one such symbol.

 

It's an evolution of the older multiphase modulation (with constant amplitude) and multilevel modulation (with constant phase). A constellation transmits about twice as many bits per symbol as them fro the same signal-to-noise ratio and availalbe band. For instance phone line modems used constellations to transmit 56kb/s on a 300-3400Hz band.

 

Equal phase spacing would be a primitive embodiment. Presently, the possible symbol values are represented in the complex plane, where the components tell the in-phase and in-quadrature ampliudes of the modulated carrier. From such a diagram, it's clear that well-spaced symbols, which give uniform and better performance, result from hexagonal or at least square pattern, not from equal phase spacing.

http://en.wikipedia.org/wiki/Constellation_diagram(actual constellations are much bigger)

 

The transitions between the successive symbols are smooth in order to stay cleanly within the permitted bandwidth (or as a result of the subsequent filters), so the phase doesn't hop between the symbols. In that sense, a phase modulation is also a frequency modulation. Varied modulation schemes exist; some do indeed impose a frequency chosen among a set so that the phase is the expected one at the end of the symbol; more elaborate schemes make an erfc transition between the phases, corresponding to a gaussian frequency swing.

 

I consider the transition details are less important, because the digital modulation alone isn't so accurate as to suppress the out-of-band spurious well enough, so an analog filter is necessary anyway, and this one will impose the nature of the transition. Anyway, this subject is of interest for specialists.

 

Old radiocomm engineers may be puzzled: just for constellation modulation, the transmitters have a modulation feedback, because the I&Q modulator wouldn't be accurate enough, and the power amplifier even less so. The output power is measured, with I and Q components, right before the antenna, is compared with what the constellation would prescribe, and the modulator gets a feedback. All that being stable.

 

Also interesting, constellations are one reason why radiocomm engineers care about phase noise. It's already important for many transmissions with synchronous detection, and gets more critical when the scheme uses symbols with smaller phase differences.

 

And because of the log, and tiny loss in throughput means a big loss in sensitivity, so the usual error-correcting codes actually lose sensitivity on constellations because of the coding (in)efficiency. Some special codes are said to gain slightly.

[John Cuthber]

"Combined amplitude and frequency modulation is used all the time."

"More precisely, a constellation modulates simulatenously the amplitude and the phase"

"In that sense, a phase modulation is also a frequency modulation"

 

Sure, and in that a carrier amplitude modulated by a sine wave gives the carrier with two additional frequencies (the sum and difference frequencies),

amplitude modulation is also a frequency modulation.

So there's no difference to discuss.

That's my point; an antenna will pick up a wiggly voltage.

How you interpret that voltage depends on how you think the signal is being transmitted.

If you try to do conventional AM and FM at the same time they will mess each other up.

OK, you can do PSK and if there's enough S/N available you can add AM.

But you are doing a totally different scheme.

It's not AM and FM at the same time (at least not in the sense of the OP's question).

 

Of course, from another pedantic point of view, if you take an FM signal, and broadcast it from an antenna then, unless that antenna has a precisely flat transmission response (and none does), the signal will always be amplitude modulated too.

All FM signals are also AM.

[Mordred]

The problem I see missing deals with distance of travel and penetration. Lower frequencies such as the AM band travel further, however if I recall higher frequencies have a better penetration depth.

 

I've worked in a radio company before. I've seen situations where the Carrier is detected but not the privacy talk. Due to different penetration values.

Although one can develop a scheme of a combination of AM, FM and phase modulations. Circuitry architect will allow it. What are the trade offs. In terms of distance, detectability over distance, detection vs penetration?

( Key note, Is strong development in this scheme worth it... The US is already tightening down its Carrier frequencies in bandwidth to try and maximize flexibility, signal strength is also limitted.). In Canada all radio frequencies must comply to FCC regulations. Don't know the US governing body, but do know Canada follows the US on radio frequencies legislation approx 5 years later on average.

One other key factor is sound quality, there is a noticeable difference between am. Fm and digital protocals

(Try listening to DTMF, compared to analog FM.)

Edited May 4, 2015 by Mordred

[Enthalpy]

JC: the phase or frequency modulation of a constellation isn't just the result of amplitude modulation. The I&Q modulator (and upstream, the digital electronics) does decide a phase for the symbol (often enough, through a frequency swing that achieves this phase near the end of the symbol) and also an amplitude for the symbol. It's really modulating both the amplitude and the phase to transmit more information - and even, modulate both with fine steps.

 

Mordred: it's not so much a matter of lower frequencies penetrating better, because the information resides very close around the carrier frequency, so you get them all with the same attenuation. It's more that a receiver has more time to identify a steady carrier (especially in AM) than a modulation that evolves quickly, and times means sensitivity. Though, this depends much on the modulation scheme, and finding the carrier and the symbol synchronization is a weakness of many modulation schemes and a hard point for the receivers. Digital receivers often rely on a special synchronization sequence contained in the message, and locking on this sequence in a reasonable time isn't trivial.