2

According to wikipedia $\mu$-law is: $$F(x) = \operatorname{sgn}(x)\frac{\ln(1+\mu|x|)}{\ln(1+\mu)} \qquad -1\le x \le 1$$

where $μ = 255$ ($8$ bits) in the North American and Japanese standards (This is a signal processing algorithm).

What operation is being performed in the $\mu|x|$ portion of the above algorithm?

Michael Hardy
  • 1
Chris Tarn
  • 121
  • 1
    Do you mean, what does $|x|$ represent? It's the absolute value of $x$ e.g. $|-1/2|=|1/2|=1/2$. – Semiclassical Sep 11 '16 at 01:20
  • 2
    suppose $x =\frac{127}{256} \approx 0.5$, $F(x)= \frac{log(128)}{log(256)}=0.875$. Now, if you consider that in this context 0.875 is better than 0.5 ... –  Sep 11 '16 at 02:46
  • 1
    I do understand $|x|$ is the abs(x). What I don't get is that $\mu$. Is this simply equal to 255 * abs(x)? – Chris Tarn Sep 11 '16 at 03:14

0 Answers0