Sound cancelling technology

I’ve research sound reduction technologies and would like to know more.
Passive sound reduction tech turns wave energy into movement or heat, right? Is the sawtooth design of anechoic chambers the best one? Are there others that do a better job in particular use cases?
The most prominent active sound reduction tech is active noise-cancelling system (“ANCS”).

  1. Delay:
    How can ANCS analyse and recreate the incoming sound fast enough to cancel it? If sound were 1000 times faster, would ANCS cease being effective?

  2. Multifrequency:
    2.1) Does ANCS only work if the sound frequency doesn’t change much?
    2.2) Could ANCS effectively cancel an incoming sound that changed frequency quickly and randomly?
    2.3) What if the incoming sound is composed of several frequencies at the same time?

  3. Could an ANCS onboard a car emit sounds on the same frequencies & amplitudes as the car but opposed in phase to enable the car to travel silently?

  4. Besides opposite-phase sound reduction, what other active sound reduction technologies exist?

I’ve read that opposite-phase active sound reduction works best with low-frequency sounds when it comes to headphones, possibly because of the tunnel-like shape of the headphones.

In an open-air environment, would opposite-phase active sound reduction also work better against low frequencies than high frequencies?

I once worked at Bose headquarters where most of the major ground-breaking development on Active Noise Cancelling technology was done (and then packaged in consumer, aviation and military headsets). I didn’t work with that technology specifically but I did get to work with some of the engineers refining it and I even won a pair in holiday raffle.

The answer to your questions are that Active Noise Cancelling technology works well but only under very specific and controlled conditions which the headphones physically provide. They don’t analyze most of the sound to be cancelled out in real-time and I am not sure that is even physically possible. Instead, they analyze the background sound over a few seconds and progressively increase the opposite sound wave until it reaches the desired results. You can hear it yourself if you ever put on a quality pair (and I mean something like Bose, not a cheap pair because lots of the knockoffs do little to nothing). You can hear the background sound fade out rather quickly but it still takes some time.

Low frequency sounds do work best for Active Noise Cancelling but they are, luckily, the sounds that are most desirable to cancel out as well. There is a surprising amount of low frequency sound all around in even “quiet” environments like a library and a whole lot more in airplanes.

Acute sounds like voices aren’t usually affected except by the passive sound cancellation of the headphone insulation itself. That is desirable for many applications because many of them are designed for environments in which users want their background noise to work but still be able to hear others talking like in aviation settings. I have never seen the technology even attempt to cancel out sounds like voices, random bangs etc. and I am not sure that is even possible either.

You also asked about open air Active Noise Canceling and I don’t believe that works very well except in the most controlled environments. I know it has been attempted in certain cars and small rooms but that is much harder to do than through headphones and won’t work at all unless you have really good control over the position of the listener and it is customized for specific sources and types of sounds. I don’t believe anyone has found a way to make it work on a large scale like stadiums or normal houses.

Yes.. Can’t remember which car maker but one of them actually added engine noise back in for the sporty sound.

I found this interesting datum:
“2015 - During the Eurovision Song Contest 2015, boos were reported from the live audience inside Wiener Stadthalle whenever a country awarded points to Russia’s entry, “A Million Voices”, which was performed by Polina Gagarina. However, anti-booing technology using active noise control[citation needed]was installed inside the arena to prevent boos from being heard on television. It is scheduled to be used in future Eurovision events.”

As you say they do have good control over the position of the sound sensor but I’m not sure how much one can call the Eurovision arena a well-controlled environment.

From what I know of it, Bose is high-quality stuff. However, each set must be affordable to an enthusiast or professional. Let’s say money is no object both for research and the marginal cost of each sound cancelling system.

If not in real-time, could the cancelling delay be brought below a second?

Are there other possible, even if expensive or complicated or merely theoretical, active ways to reduce the amplitude of a sound that makes it to a sound sensor/listener?

Note that I only say “reduce” and not completely cancel out; Even a 50% reduction in what makes it back to the listener/sound sensor could be a major benefit in some situations.

Finally, please humor me here about the following question:

Let’s say I’m writing a fantasy book and I want it to be a hard science fantasy book when it comes to the technology used by the characters.

In my book, the hero must venture into Vladland. Vladland is inhabited by vampires. The vampires are very nearsighted so they rely on their friends the bats to warn them of intruders and tell them where the intruders are. Those bats, as bats do, emit sound waves in varying frequency chirps to detect and locate intruders. When the bats think there is an intruder present, they sometimes change the frequencies of the chirps they emit.

If the hero has extremely fancy (but still theoretically possible) technology, could he actively reduce the sound that makes it back to the bats fast and well enough to have a chance to survive in Vladland?

I have a set of Bose QC20 earbuds and - testing right now - they do a pretty good job of silencing the football announcer on TV (comparing with the switch on/off, earbuds in for both). It’s not perfect, but the sound goes from 100% intelligible to an indistinct whisper. I don’t know if it’s specifically canceling the voices or not, but it’s getting the job done.

I just tried a pair of NC Bose headphones last week in a Bose store that had a feature that *enhanced *voices for just such a purpose.

There’s some evidence that some species of tiger moth can do this in nature. They emit ultrasonic clicks to “jam” the bat’s sonar. Wiki link. In experimental conditions the bats had a 10 fold decrease in success at catching the jamming moths vs silent ones.

I suspect though that the jamming is only effective at making it harder for the bats to tell where the moths are rather than being unable to detect them at all, but I think the same principle can be applied to your hero.


Thanks for that. I had no idea that jamming was used by animals.

What I’m talking about though is active sound reduction rather than what the moths seem to be doing. Those moths are doing the equivalent of shining a spotlight in the eyes of their predators which has definite drawbacks for the moths.

My hero is looking to reduce the amplitude of the echo that is heard by the bats which is much safer for him than what the moths are doing.

Do I understand correctly that my hero could only use opposite-phase active sound reduction if the bats were sending all their calls at the same frequencies, the same frequency modulations and at the same intervals? Or at least, freqs & intervals my hero could predict before the bats sent them?

When using active noise-cancelling headphones, if the background noise shifts in frequency, does the headphone have to start the noise cancelling all over again or might it have a chance to follow the frequency change of the noise?
The Bose website mentions their noise-cancelling headphones being effective in crowded city streets and busy offices. Are they just overselling their product?

If not, then that technology would have to be able to shift frequencies rather quickly since sudden noises occur in busy offices (e.g.: ringing phones) and both sudden noises and frequency shifts occur in crowded city streets (e.g.: talking, cars moving toward/away).

Personal experience: I own a pair of Bose QuietComfort 20i in-ear noise cancelling headphones. They are very good at cancelling steady noises like fans, refrigerator compressors, rolling noise in the train. Moderately good at cancelling monotonous, droning (nonexcited-male-type) conversation. Unfortunately almost useless at cancelling sudden, high pitched shrieks (my female coworkers).

What about 1) sudden, low-pitched noises 2) steady, high-pitched noises 3) sounds that start low and then increase in pitch?

Isn’t what you’re looking for the same as the stealth technology used by military aircraft except applied to sound instead of radar? In that case your hero could just wear a suit of that anechoic chamber material which would either absorb or minimally reflect the pings from the bats.

You’re missing something fundamental about the physics here.

There is no way, even in principle, for a jammer at location B to emit a wave that cancels the signal from an emitter A as seen/heard by receivers at all potential locations C through Z.

At any given instant the two waves from A & B will create an interference pattern in 3-space. At some points there will be zero amplitude, at others there will be double or more amplitude. Depending on where the receiver C is sitting in that field at that same instant they’ll receive either that zero, or that double+, or any value in between.
In principle, if jammer B had absolute knowledge of A’s location and near-future trajectory, and C’s location and near-future trajectory, and what signal train A would emit in the near future, all out to the time of flight of the jamming signal from B to C (plus the time required to compute all this), THEN, and only then, in principle could B construct a jamming signal that would arrive at C ideally crafted to cancel the signal A will emit to arrive at the same time.

But while doing all that, any other receiver at location D (e.g. C’s wingman) will not have the same result. D will see something other than zero. D will see a signal from A and a signal from B.
Make A represent engine noise, B represent magic ANR tech, and C & D represent two passengers in the car with 4 ears, and you see why car- or room-scale ANR via phase cancelation is not possible. There are some ANR schemes which do work, but it’s essentially barrage noise jamming that works in the psychological channel of the human hearing system.

Based on the “citation needed”, I think what really happened was this:

Eurovision had a DSP in the audio recording path. That processed the signal after it was simple L+R stereo on the way to the final audio recording. It’s pretty easy to dynamically filter out unwanted noises in that domain. It’s also pretty easy to add laugh track or cheer track to further drown out undesirable sounds such as boos.

IMO the wiki author jumped to the unwarranted (and unsupported by cite) conclusion that the booing was somehow cancelled in the room via acoustic means.

Bottom line IMO: everybody in the room heard all the booing. They almost had to, since booing is a social phenomenon; most people only do it when they hear the rest of the crowd doing it. And if it was being magically cancelled, most people wouldn’t hear anybody else doing it and therefore wouldn’t be moved to do it themselves. And therefore there’d be almost nothing which needed to be cancelled on the TV broadcast.

As musicians have been saying for 70+ years now: “Don’t worry, we’ll fix it in the mix.”

As has been described, there is no mechanism for creating a sound field from a speaker that will cancel out what you want by simply having a system try to emit an out of phase signal.

Insanely fancy, it is possible. If all you want is a theoretical capability, rather than anything that is much more than science fiction, but sci-fi that doesn’t break any rules of physics, well OK. The trick is, in part, the concentrate on cancelling at the reflection.

You could imagine a suit that adsorbs all sound in the wavelengths used by the bats. This is hard, but within reason since the wavelengths are reasonably short. 34kHz is 10mm. However your moth hunting bats apparently go down to 12.7kHz, which is more like 25mm. This matters as the higher the frequency the easier it is to make sound adsorbing material. In reality a passive absorber suit might be possible with modern materials. It would probably be a bit hot to wear. A very thick suit made of low density felted microfibre would probably work. What you need to absorb sound is very high surface area per unit volume, which is why you use micro-fibre.

An active surface that reacts to the incoming sound wave, and essentially eats the wave is the silly tech version. You are creating an active termination to the air. You can wave your hands and talk about nano-tech. A suit covered with micro-machined actuators that feel the vibrations in the air and damp any energy hitting them that is within the frequency of the bat’s radar.

The anechoic ECM you mention is potentially quite effective against high freqs, much less so against low freqs, especially for less-than-bomber-size aircraft. Whereas opposite-phase cancelling is most effective against low freqs. So you can see why it would be interesting to combine them.

Additionally, anechoic ECM is a prissy princess to maintain, from what I’ve read.


Fundamental physics incoming…

Sure, I should have excluded that possibility and been more explicit when I posed the question. Multistatic radar networks are nifty.

Ah, yes. Against air assets, which are usually close enough to each other to catch the same echo but far enough apart that the echo will look different, ANR wouldn’t be much use. In that situation, the anechoic-ECM would often be best, unless up against an E-2 Hawkeye-like system.

What about this situation: An aircraft with anechoic chamber-like ECM reduces the detection range of enemy radars such that there is no detection overlap between enemy radars zones. When the aircraft gets close enough to the enemy ground/surface radar which is its target, the aircraft detects the radar’s emission and starts using ANR against that radar to further reduce its range until it’s close enough to launch its payload at the radar or relay targeting info to another platform.

Could that be effective?
I can see that one countermeasure against that ANR tactic would be to have 2 or more ground/surface radar receivers in close-but-slightly-different locations so that even with anechoic ECM, their detection zones would overlap. That would require a near doubling of enemy radar resources though, wouldn’t it?

This is what I initially wondered about. Is it possible to compute it so fast? Is ANR limited to radars that aren’t switching up their freqs and PRF?

Maybe if the jammer captures the radar’s gates then starts using ANR?

Thanks for that explanation.

The best passive sound reduction us an air-tight seal, combined with material near the seals that don’t pick up and transmit the vibrations from the air on the noisy side.

I use my QC20s in crowed bars and it cuts out the background noise enough for me to clearly hear the person or group I am sitting with. They have a setting that changes the amount of cancellation so that voices and background noise come in a little clearer. They recommend that setting for people jogging so that they can hear cars and people on biles a little more clearly. It’s pretty cool.

I have both the Bose ear buds and much lower tech jobsite ‘electronic’ hearing protection. The jobsite ones are really just good quality PPE ear muffs with active Mic’s that allow you to hear what’s going on around you minus high frequencies. The Mic’s cut out if there are any loud noises. They are great around a shop. The Bose ear buds are a god send on flights.