How to make an AM transmitter?

[rfblackmagic]

I want to verify if the silent subliminal made with this script in Audacity is working as it should.

The script:

;nyquist plug-in
;version 1
;type process
;name "Subliminal..."
;action "Subliminal..."
;control carrier "Carrier" real "Hz" 17500 14000 20000

(setf carrier (max 14000 (min carrier 20000)))

;; We have two Nyquist frequencies, carrier/2 and *sound-srate*/2.
;; The CUTOFF is the maximum allowed frequency in the modulator.
;; It must not be greater than carrier/2, but also not greater than
;; the difference between the carrier and *sound-srate*/2, because
;; otherwise the modulated carrier aliases.

(setf cutoff (min (/ carrier 2.0) (- (/ *sound-srate* 2.0) carrier)))

(defun cut (function sound frequency)
  (dotimes (ignore 10 sound)
    (setf sound (funcall function sound frequency))))

(defun subliminal (sound)
  (let ((result (mult 2 (cut #'lowpass8 (hp sound 80) cutoff)
                        (hzosc carrier))))
    (cut #'highpass8 result carrier)))

(if (< *sound-srate* 44100)
    (princ "The track sample frequency must be minimum 44100Hz.")
    (multichan-expand #'subliminal s))

The script claims to generate an SSB AM signal, in which the voice frequencies are upshifted to near the threshold of audible frequencies, humans can hear between 20Hz to 20KHz but as they get older, they lose sensitivity towards 20KHz, so the script shifts the voice frequencies towards these higher frequencies. The theory is that when this is played, because the conscious mind doesn't hear it, it directly goes the subconscious mind.

So I tried that script on few messages, but I wasn't able to notice any difference in my thought process.

But the problem with the script is it amplitude modulates the messages rather than PCM in which most audio files are stored, so I want to know what kind of output would I get if I transmitted an audio file generated from this script on AM. Is there any transmitter I can build to verify if it is producing the output it should.

 

[Nigel Goodwin]

Apart from the fact the entire premise is almost certainly wrong, AM radio is only low frequency, (even FM or DAB is too low) and the high frequencies won't pass through it.

 

[danadak]

I want to verify if the silent subliminal made with this script in Audacity is working as it should.

The script:

;nyquist plug-in
;version 1
;type process
;name "Subliminal..."
;action "Subliminal..."
;control carrier "Carrier" real "Hz" 17500 14000 20000

(setf carrier (max 14000 (min carrier 20000)))

;; We have two Nyquist frequencies, carrier/2 and *sound-srate*/2.
;; The CUTOFF is the maximum allowed frequency in the modulator.
;; It must not be greater than carrier/2, but also not greater than
;; the difference between the carrier and *sound-srate*/2, because
;; otherwise the modulated carrier aliases.

(setf cutoff (min (/ carrier 2.0) (- (/ *sound-srate* 2.0) carrier)))

(defun cut (function sound frequency)
  (dotimes (ignore 10 sound)
    (setf sound (funcall function sound frequency))))

(defun subliminal (sound)
  (let ((result (mult 2 (cut #'lowpass8 (hp sound 80) cutoff)
                        (hzosc carrier))))
    (cut #'highpass8 result carrier)))

(if (< *sound-srate* 44100)
    (princ "The track sample frequency must be minimum 44100Hz.")
    (multichan-expand #'subliminal s))

The script claims to generate an SSB AM signal, in which the voice frequencies are upshifted to near the threshold of audible frequencies, humans can hear between 20Hz to 20KHz but as they get older, they lose sensitivity towards 20KHz, so the script shifts the voice frequencies towards these higher frequencies. The theory is that when this is played, because the conscious mind doesn't hear it, it directly goes the subconscious mind.

So I tried that script on few messages, but I wasn't able to notice any difference in my thought process.

But the problem with the script is it amplitude modulates the messages rather than PCM in which most audio files are stored, so I want to know what kind of output would I get if I transmitted an audio file generated from this script on AM. Is there any transmitter I can build to verify if it is producing the output it should.

Hearing loss causes :

Age-Related Hearing Loss (Presbycusis) — Causes and Treatment

Age-related hearing loss (also called presbycusis, pronounced prez-buh-KYOO-sis) is hearing loss that occurs gradually for many of us as we grow older.

www.nidcd.nih.gov

So what are the script authors saying leads to their conclusion ? I can see if the tympanic membrane
hardens that would compromise, hence shifting more power to the higher freqs would enhance.
After all DSPs are used now in hearing aids to shape amplitude versus frequency.


Regards, Dana.

 

[rjenkinsgb]

The script claims to generate an SSB AM signal, in which the voice frequencies are upshifted

All it is doing, if it works, is frequency shifting the input.

That is what a normal "SSB" transmission is (using upper sideband, anyway) - the literal audio, shifted so it's original zero frequency reference is the nominal RF carrier frequency; or just up a few thousand, hundreds or tens of Hz, for voice effects or feedback suppression.

Audacity's "Change pitch" function should do exactly the same thing.

You could change pitch downward to restore the original audio.

[Lower sideband inverts or mirrors the audio frequency spectrum, so the highest input frequency is the lowest output frequency].

 

[rfblackmagic]

Apart from the fact the entire premise is almost certainly wrong, AM radio is only low frequency, (even FM or DAB is too low) and the high frequencies won't pass through it.

I don't think the entire premise is completely wrong, it depends on how the cochlear resonates to these sounds and if signals fire to the auditory cortex. AM is in KHz range, has enough bandwidth for voice at least. I think you are confusing radio waves with sound waves. 20KHz in sound waves is different 20KHz in radio waves.

Hearing loss causes :

Age-Related Hearing Loss (Presbycusis) — Causes and Treatment

Age-related hearing loss (also called presbycusis, pronounced prez-buh-KYOO-sis) is hearing loss that occurs gradually for many of us as we grow older.

www.nidcd.nih.gov

So what are the script authors saying leads to their conclusion ? I can see if the tympanic membrane
hardens that would compromise, hence shifting more power to the higher freqs would enhance.
After all DSPs are used now in hearing aids to shape amplitude versus frequency.


Regards, Dana.

It is actually based on this patent: https://patents.google.com/patent/US5159703

As you can see the original one asks for some electronic components. The script author implemented it in the Nyquist script in Audacity, according to the script author the output would be mathematically identical to the output obtained by the method described in the patent.

All it is doing, if it works, is frequency shifting the input.

That is what a normal "SSB" transmission is (using upper sideband, anyway) - the literal audio, shifted so it's original zero frequency reference is the nominal RF carrier frequency; or just up a few thousand, hundreds or tens of Hz, for voice effects or feedback suppression.

Audacity's "Change pitch" function should do exactly the same thing.

You could change pitch downward to restore the original audio.

[Lower sideband inverts or mirrors the audio frequency spectrum, so the highest input frequency is the lowest output frequency].

Although I simplified it as frequency shifting, it is doing quite more than that, I don't know Nyquist, so I don't know what the lines are doing, an electronic version of the script is described in the patent I linked above in my comment to Dana. First a sine tone at around 400KHz is modulated by a message, this in turn is added to another sine tone at 400KHz, the difference of these two sine tones contains the message at around 14.5KHz audio frequency.

 

[Nigel Goodwin]

I don't think the entire premise is completely wrong, it depends on how the cochlear resonates to these sounds and if signals fire to the auditory cortex. AM is in KHz range, has enough bandwidth for voice at least. I think you are confusing radio waves with sound waves. 20KHz in sound waves is different 20KHz in radio waves.

No, you are confusing what AM radio is - the AUDIO bandwidth is very low, as you say it has voice bandwidth (and a bit more actually) but nothing like 20KHz bandwidth. The highest frequency is only about 5KHz, well audible to almost everyone.

It is actually based on this patent: https://patents.google.com/patent/US5159703

A patent means very little, there are millions of patents that don't work, particularly in America, although in more modern times they have tightened up on what they will issue a patent for.

However, at least the patent doesn't seem to claim it can be transmitted over am radio.

 

[rjenkinsgb]

First a sine tone at around 400KHz is modulated by a message, this in turn is added to another sine tone at 400KHz, the difference of these two sine tones contains the message at around 14.5KHz audio frequency.

That's common method of frequency shifting audio.

Mix up to a higher frequency, put that through a high or low pass filter depending on which sideband (sum or difference) is wanted, then mix with nearly the same frequency to again give a sum and difference.

The sum in that case would be around 800KHz so easily removed, the difference would be audio offset up in frequency by the difference in the two carrier frequencies.

 

[rfblackmagic]

No, you are confusing what AM radio is - the AUDIO bandwidth is very low, as you say it has voice bandwidth (and a bit more actually) but nothing like 20KHz bandwidth. The highest frequency is only about 5KHz, well audible to almost everyone.



A patent means very little, there are millions of patents that don't work, particularly in America, although in more modern times they have tightened up on what they will issue a patent for.

However, at least the patent doesn't seem to claim it can be transmitted over am radio.

What is my confusion about AM radio? The bandwidth limitation is not natural, it is artificially imposed by FCC or something like that, because the frequency range allocated to AM radio is KHz range and if anyone transmitted full range of audio frequency, they'd interfere with adjacent channels of other stations. Although in licensed in radio stations, the highest audio frequency which can be broadcasted is 5KHz, they have option to select which 5KHz part of the audible frequency range they transmit, this 5KHz could be at the lower end, comprising bass and mid-range or it could be anywhere in the treble range, which wouldn't be audible to adults but audible to children.

That's common method of frequency shifting audio.

Mix up to a higher frequency, put that through a high or low pass filter depending on which sideband (sum or difference) is wanted, then mix with nearly the same frequency to again give a sum and difference.

The sum in that case would be around 800KHz so easily removed, the difference would be audio offset up in frequency by the difference in the two carrier frequencies.

So is there a way to make an AM transmitter?

 

[rjenkinsgb]

Ignoring the frequency response and assuming it was adequate, whatever you modulate an AM transmitter with, you would get the same back, out the receiver.


SSB transmission is based around frequency mixing, often a similar two stage process as I described for audio but using a fixed radio frequency for the first stage, to do the sideband selection filtering before mixing again to the final frequency.

eg.
They call the first mixer the "SSB generator" in this diagram. The sideband filtering often at around 10.7 MHz for VHF equipment, as narrowband quartz filters are readily available for that.

Think of it as a reverse superheterodyne receiver, with the transmit signal generated at the intermediate frequency and mixed to the on-air frequency then amplified.

 

[Nigel Goodwin]

What is my confusion about AM radio? The bandwidth limitation is not natural, it is artificially imposed by FCC or something like that, because the frequency range allocated to AM radio is KHz range and if anyone transmitted full range of audio frequency, they'd interfere with adjacent channels of other stations. Although in licensed in radio stations, the highest audio frequency which can be broadcasted is 5KHz, they have option to select which 5KHz part of the audible frequency range they transmit, this 5KHz could be at the lower end, comprising bass and mid-range or it could be anywhere in the treble range, which wouldn't be audible to adults but audible to children.

You still seem totally confused about it

The transmitter, and the receiver, are both bandwidth limited to the low audio end of the spectrum, with an upper limit of 5KHz, and a lower limit presumably about 20Hz or so (but that is of no relevance).

Why on earth would you imagine you can use a 5KHz slice of bandwidth anywhere you want? - if you try to transmit 5KHz to 10KHz through it you will get next to nothing, as it can't go above 5KHz, and would also require 10KHz of bandwidth as well (if the transmitter allowed it, which it wouldn't).

 

[rfblackmagic]

You still seem totally confused about it

The transmitter, and the receiver, are both bandwidth limited to the low audio end of the spectrum, with an upper limit of 5KHz, and a lower limit presumably about 20Hz or so (but that is of no relevance).

Why on earth would you imagine you can use a 5KHz slice of bandwidth anywhere you want? - if you try to transmit 5KHz to 10KHz through it you will get next to nothing, as it can't go above 5KHz, and would also require 10KHz of bandwidth as well (if the transmitter allowed it, which it wouldn't).

You have any proof this is not a hardware limitation and a limitation of the modulation?

Are you telling me no one can build an AM transmitter which transmits the full range of audible frequencies?

 

[Kerim]

Are you telling me no one can build an AM transmitter which transmits the full range of audible frequencies?

One can do it.
But, you know, there are rules in using the radio spectrum.
Even if we forget the rules or one can get a license to occupy 20Khz bandwidth or more (by using SSB), the question is: By which AM receiver one may hear the transmitted full audio signal? It has to be, in turn, a special SSB receiver whose baseband extends to 20Khz not 5Khz for example. Is this what you have in mind?

 

[Nigel Goodwin]

You have any proof this is not a hardware limitation and a limitation of the modulation?

Are you telling me no one can build an AM transmitter which transmits the full range of audible frequencies?

No, it's a hardware limitation - you could increase bandwidth considerably, depending on the carrier frequency you use.

But as you're specifically talking about AM radio transmitters, the 5KHz limit applies - however, if you were building illegal transmitters and receivers you could design them as you wish (until you're caught by the relevant authorities).

 

[rfblackmagic]

One can do it.
But, you know, there are rules in using the radio spectrum.
Even if we forget the rules or one can get a license to occupy 20Khz bandwidth or more (by using SSB), the question is: By which AM receiver one may hear the transmitted full audio signal? It has to be, in turn, a special SSB receiver whose baseband extends to 20Khz not 5Khz for example. Is this what you have in mind?

I think I acknowledged the existence of these rules when I talked about FCC. That is what I had in mind, build an AM transmitter and receive it on AM radio, if it not possible, on an SDR. I think something like this is doable with some Arduino modules.

No, it's a hardware limitation - you could increase bandwidth considerably, depending on the carrier frequency you use.

But as you're specifically talking about AM radio transmitters, the 5KHz limit applies - however, if you were building illegal transmitters and receivers you could design them as you wish (until you're caught by the relevant authorities).

But who has the power and resources to build a transmitter which will interfere with other AM broadcasts? You certainly were arguing with me earlier as if it was a limitation imposed by modulation type. 433MHz is unlicensed, so I was thinking of using that or CB frequencies.

 

[rjenkinsgb]

Again:

If you build an AM transmitter and modulate it with an audio signal, no matter how frequency shifted, mangled or scrambled it is, then use an AM receiver to receive and demodulate that, you will just get back the original signal.


You appear to be confusing AM and SSB techniques. SSB transmission and reception allow frequency shifting by offsetting the relative tuning.

(You can receive an AM signal on an SSB receiver, but with any frequency offset still within the passband range, you get a massive heterodyne from the AM carrier).

20KHz in sound waves is different 20KHz in radio waves.

That is a fundamental misconception.

The only difference between sound and radio is with the transducers or signal medium part, whatever converts between electrical signals and physical vibration or electromagnetic wave.

Acoustic signals are used up to the MHz range in some things, and the lowest in-use radio transmitter frequency I'm aware of is 16 KHz, within the "normal" audio range.

 

[Nigel Goodwin]

I think I acknowledged the existence of these rules when I talked about FCC. That is what I had in mind, build an AM transmitter and receive it on AM radio, if it not possible, on an SDR. I think something like this is doable with some Arduino modules.

Again, you don't seem to understand? - IT WON'T WORK THROUGH A STANDARD AM RADIO

But who has the power and resources to build a transmitter which will interfere with other AM broadcasts? You certainly were arguing with me earlier as if it was a limitation imposed by modulation type. 433MHz is unlicensed, so I was thinking of using that or CB frequencies.

Both of which would be illegal, and you're likely to be certainly generating enough power to cause interference - no one ever suggested you would wipe out commercial stations, although you would if your signal is strong enough and the commercial station is week enough, such as a fair way from the commercial transmitter.

'Unlicensed' (or 'licence free') doesn't mean you can do what you want, it's just as regulated as other frequencies - and wiping out the 433MHz band or CB band is very likely to get you reported.

 

[rjenkinsgb]

433MHz is unlicensed

"Licence free" means the user does not need a government-issued licence, unlike business PMR, broadcast or Amateur radio.

It does not mean the equipment is unregulated - anything legal for use by an unlicensed person has strict manufacturing and/or usage regulations.


eg. There are very! strict limitations on the use of 433 MHz in the USA and many other countries.

For data transmission, the FCC restrictions include such as no more than five seconds transmission for manually operated transmitters such as remote controls, or two seconds transmission per hour from any one automated device, with a one second maximum duration for any single transmission burst.

The actual power limit is based on RF field strength, but roughly equivalent to 5mW into a quarter wave antenna.


Voice transmission in the 430 - 440 MHz band requires an Amateur (Ham) Radio licence in the USA, and either that or the use of commercial type-approved LPD433 radios (which have a 10mW power limit) in Europe / CEPT countries.


Note that quite a few of the people on this site are licenced radio amateurs as well as electronics engineers / designers or enthusiasts.

Any of us with the appropriate class of licence for the region we are in can legally design, build and self-certify transmitters for frequency bands throughout the radio spectrum and with power output levels that could easily wipe out commercial radio stations over large areas if mis-used..
(Anyone causing such deliberate interference would of course lose their licence and likely face serious fines etc).