Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Princen Bradley condition (Read 8035 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Princen Bradley condition

Can any1 explain me the Princen Bradley condition required for windows in MDCT
1.  h(k)^2 + h(k+N)^2 = 1

2. h(k) = h(2N-1-k)

Princen Bradley condition

Reply #1 –
Quote from: pratheekp on
Can any1 explain me the Princen Bradley condition required for windows in MDCT
1.  h(k)^2 + h(k+N)^2 = 1
2.  h(k) = h(2N-1-k)

Try implementing the MDCT in terms of the Type-IV DCT. This may help you understand the necessary window conditions to make the "whole" transform (many blocks) reversible and optionally orthogonal.

Princen Bradley condition

Reply #2 –
One more MDCT related question. MDCT is a lapped transform and psychoacoustic models rather use FFT (both 'even' and 'odd' components included). Wouldn't it be right to adjust masking thresholds using prior knowledge of quantization errors in previous frame?

Princen Bradley condition

Reply #3 –
Any ideas? I believe either Sebastian or Chris Helmrich should have thought about it..

Princen Bradley condition

Reply #5 –
Quote from: pratheekp on
Can any1 explain me the Princen Bradley condition required for windows in MDCT
1.  h(k)^2 + h(k+N)^2 = 1

2. h(k) = h(2N-1-k)


and not
h(k) = h(2N-k) instead of h(2N-1-k) ?
 

Princen Bradley condition

Reply #6 –
No, this just ensures symmetry and k starts with 0. (see n in my post here)
"I hear it when I see it."