Loading [MathJax]/jax/output/HTML-CSS/fonts/TeX/fontdata.js

Exercise 1.1: Music Signals

From LNTwww
Revision as of 13:26, 22 December 2020 by Noah (talk | contribs)

Music signals, original,
noisy and/or distorted?

On the right you see a ca.  30 ms  long section of a music signal  q(t). It is the piece „For Elise” by Ludwig van Beethoven.

  • Underneath are drawn two sink signals  v1(t)  and  v2(t), which were recorded after the transmission of the music signal  q(t)  over two different channels.
  • The following controls allow you to listen to the first fourteen seconds of each of the three audio signals  q(t)v1(t)  and  v2(t).


Original signal  q(t)

Sink signal  v1(t)

Sink signal  v2(t)



Notes:



Questions

1

Estimate the signal frequency of  q(t)  in the displayed section.

The signal frequency is approximately  f=250Hz.
The signal frequency is approximately  f=500Hz.
The signal frequency is about  f=1kHz.

2

Which statements are true for the signal  v1(t) ?

The signal  v1(t)  is undistorted compared to q(t).
The signal  v1(t)  shows distortions compared to  q(t) .
The signal  v1(t)  is noisy compared to  q(t) .

3

Which statements are true for the signal  v2(t) ?

The signal  v2(t)  is undistorted compared to  q(t) .
The signal  v2(t)  shows distortions compared to  q(t) .
The signal  v2(t)  is noisy compared to  q(t) .

4

One of the signals is undistorted and not noisy compared to the original   q(t) .
Estimate the attenuation factor and the running time for this.

α = 

τ = 

 ms


Solution

(1)  Correct is the solution 2:

  • In the marked range of 20 milliseconds approx.   10  oscillations can be detected.
  • From this the result  follows approximately for the signal frequency; f=10/(20ms)=500Hz.


(2)  Correct is the solution 1:

  • The signal  v1(t)  is undistorted compared to the original signal q(t). The following applies:   v1(t)=αq(tτ).
  • An attenuation  α  and a delay  τ  do not cause distortion, but the signal is then only quieter and comes later than the original.


(3)  Correct are the solutions 1 and 3:

  • One can recognize both in the displayed signal  v2(t)  and in the audio signal  additive noise   ⇒   solution 3.
  • The signal-to-noise ratio is approx.   30 dB; but this cannot be seen from this representation.
  • Correct is also the solution 1:   Without this noise component  v2(t)  identical with  q(t).


(4)  The signal  v1(t)  is identical in form to the original signal  q(t)  and differs from it only

  • by the attenuation factor  α=0.3_  (dies entspricht etwa  –10 dB)
  • and the delay  τ=10ms_.