I am currently working on ultrasonic frequencies (20 kHz-60 kHz). According to a news snippet I found:

The state government has also tried employing ultrasonic guns that can scare monkeys away and create van vatika or small sanctuaries for different troops. The plan to use ultrasonic guns, which cost Rs 20,000 each, was shelved after municipal officials said the ultrasonic frequencies jammed phone and internet services.

stated that the Ultrasonic Guns jammed up phone and internet services.

Aren't Ultrasonic (Mechanical energy) and Phone / Internet transmissions (Electromagnetic energy) completely different in nature and operate on different spectrum? Ultrasonic waves they might have used might go up to 100 kHz, while the Internet (2.4 GHz) and mobile systems (GSM: 900 MHz and above) are in completely different spectrum. Is this possible?

  • 1
    $\begingroup$ by the way, I always get itching eye when I see someone write "KHz" and "Ghz" in the same question: It's a small "k" for "kilo", (not "K", because that'd be "Kelvin"), and a capital "H" in "Hz" for "Hertz" (not "hz", because that would be hours·z), so it's kHz and GHz. I know this is nitpicking, because we all know what you've meant, but it's not really good style to mix these up :) $\endgroup$ Jul 16, 2018 at 11:34
  • $\begingroup$ By the way, 20,000 INR would be an incredibly cheap weapon device. (I've converted that to Euros because I was curious, and it's 250€) That gives @hotpaw2's answer extra credibility: Either the price was incorrectly reported (because someone omitted the R&D costs or something similar), or these are really hastily designed and then, bad EMI symptoms are thoroughly believable. $\endgroup$ Jul 16, 2018 at 11:37
  • $\begingroup$ @MarcusMüller. Edited it. I'm afraid I never realized how wrong my question looked from a scientifically expressive point of view. $\endgroup$
    – Mahip
    Jul 16, 2018 at 12:41
  • $\begingroup$ as said, really just nitpicking, don't worry :) $\endgroup$ Jul 16, 2018 at 12:42

2 Answers 2


Not sure about up into the GHz ISM bands, but unfiltered, unshielded switching power amplifiers and power supplies running at high audio frequencies can generate and radiate harmonics well into the RF HF bands.

  • $\begingroup$ Oh, okay. On a circuit-level, this is possible. But any idea on how/if directed Audio signals can induce such an interference on the amplifiers? $\endgroup$
    – Mahip
    Jul 16, 2018 at 6:52
  • $\begingroup$ These high-power directed audio signals are usually generated using electronic amplifiers of various sorts. $\endgroup$
    – hotpaw2
    Jul 16, 2018 at 16:44
  • $\begingroup$ Then that must be it. Since they haven't provided any technical details in the news, this seems like a possible answer. $\endgroup$
    – Mahip
    Jul 17, 2018 at 1:52

You're right, acoustics is temporal change of air pressure, whereas radio is electromagnetic waves. These are independent phenomena.

However, electromagnetic waves do behave differently based on the medium they travel: Water, for example, has a different refractive index than air for the electromagnetic wavelengths we consider visible, which is why objects look like they bend when entering water. Same applies to radio waves (which is just "light with a much lower frequency") e.g. when entering a PTFE lens or when going through parts of the atmosphere with more charged particles in it.

However, you (without extremely involved lab equipment at least) won't be able to even measure the difference between "high-pressure air" and "low-pressure air" as caused here. Even less so, since in sound, these areas will be of the same size and "average out".

One can, however, use the fact that the medium moves, for Doppler-based wind measurement. Again, however, the net movement of air will be very small.

So, "in the air", no significant modification of your radio waveform will happen.

"In your electronics" is a completely different thing, though. Ceramic capacitors, especially high-density ones, are subject to microphonic effects. These are used everywhere, for example in the voltage stabilization of RF amplifiers inside your phone (though potentially indirectly).

If you charge a capacitor up to certain voltage, and then disconnect it, it will hold that voltage. If you then measure the voltage, and apply vibrations to the capacitor, you'll see these vibrations in the voltage across it. It's a well-known effect, and there's even types of microphones that are based on this principle.

Now, if you're building sensitive measurement equipment, you'll do your best to shield it from such vibrations. In a smartphone, you've got no chance whatsoever, and have to live with them.

For the RF part, it's really a question of whether that matters at all: At RF signal bandwidths in the order of 10 MHz, e.g. an amplifier gain control loop will probably deal well with a couple kHz of low-amplitude oscillation in the supply voltage.

Now, interesting is what happens on the audio end: Whereas I'll wildly assume that you still won't hear the fact that maybe audio coming from your speaker also contains a bit of downmixed ultrasonic sound (if your own ears can't hear it, can't be that bad), it's a well-known fact that due to imperfections in the anti-alias filtering that is possible within the constraints of analog components in a phone, you'll see aliases of your 60 – 90 kHz oscillation in your audio signal after the ADC: There's attacks on smart homes where people just mix up a recording of "Hey Alexa, order two metric tons of toilet paper. Confirm." mixed up with the sampling rate of the ADC, which leads to the device "hearing" the sound as if it was in the audible range. In the case of your acoustic weapons, it'd probably sound like a very annoying siren or beep or loud noise.

  • $\begingroup$ I also think that the ultrasound leakage into the voice encoder might be the issue... $\endgroup$
    – Fat32
    Jul 16, 2018 at 21:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.