# Why noise canceling headphones do not attenuate the signal itself?

Noise canceling headphones use a microphone to capture noise in order to cancel the same. The microphone will capture whatever acoustic signal reaches it. Since it is in the vicinity of the speaker, it must capture the sound coming from the speaker apart from the noise, particularly low frequency components. This should result in not only removal of noise but also of some portions of the signal. How it is ensured that only noise is removed?

It's not clear what you are asking. Noise cancelling headphones measure the noise to be cancelled, then actively produce the negative of that noise at just the right volume so that the two cancel.

There is no issue of "recording" anything here. The microphone data is used real time, with perhaps a tiny delay to account for the propagation of the waves it measured across the distance to the emitter. The speed of sound is roughly 3 µs/mm, so even if the mic and speaker were 10 mm apart (a lot), then that would only be 30 µs of "recording" before the recorded information would be discarded. In practise, this would only require a single sample anyway.

As for "attenuate the same", it is unclear what you are talking about.

A better answer requires a better question.

• Please see the edited question, hopefully it is better now. – akm Jun 27 '16 at 1:56

Consider that the noise is being 'cancelled' by your perception because of the speaker in your ear. Most headphones of this type also have some sort of material to absorb vibrations so that the microphone couldn't pick up any sound coming from the speaker.

If the microphone could somehow collect sound from the speaker, the antiphase sound through the speaker would be out of phase with the original sound that was picked up by the microphone, and the microphone would then not pick up any background sound. The computer in the headphones would not produce any antiphase sounds, then the microphone would pick up the ambient noise again in the next iteration. This would simply result in very poor quality headphones that would probably hardly cut any ambient noise at all by human perception.

That being said, the speaker in any headphone that does not have some sort of noise absorption material (passive noise cancelling) is going to be far enough away from the microphone that the microphone couldn't pick up the sound. You shouldn't be listening to anything greater than 80dB for a prolonged period of time, and the inverse square law

$$\frac{I}{{I}'}= \left ( \frac{{d}'}{d} \right )^{2}$$

Shows pretty clearly that simply the distance between the speaker and the microphone isn't enough to reduce the dB at the microphone substantially.

## Simply

It would be terrible headphone design, so they don't make them that way. They use noise absorbing material such as padding on over-ear headphones and rubber on earbuds.

Noise cancelling headphones work by measuring external nose, inverting the signal produced and adding it to the speaker output. This either produced a net effect of 'quiet' or eliminates noise from the audio feed. Separating and audio input from external noise is an inherent part of this and can be achieved through microphone placement, acoustic insulation, electronic/digital filtering or any combination of the above.

There is also the fact if the system 'knows' what the desired audio output is it can be added after any nose cancellation takes place and it is also possible to predict what frequencies are likely to be problematical in terms of noise in advance, depending on context.

But the simple answer is that the whole system is designed so that the external noise and desired audio output are separated.

Noise cancellation tends to work less well for low and mid frequencies, apart form anything else because there is often a residual need or verbal communication and allowing audible warnings to work, especially in an industrial context. But having said that it is often high frequency noise which is most annoying and hazardous in industrial applications to there is a lot of scope for specific tuning with filters and parametric and graphic EQ especially where software allows this to adapt to ambient conditions.

• Microphone placement and insulation will not keep away the low frequency components. How electronic filtering works? – akm Jun 27 '16 at 1:58