User`s guide

3 Working with Signals
3-48
•Output buffer size (per channel), M
o
•Buffer overlap, L
•
Initial conditions
Buffering an N-channel (1-by-N or N-by-1) sample-based signal produces a
M
o
-by-N frame-based signal. Buffering an M
i
-by-N frame-based signal (N
channels and M
i
samples per frame) results in an M
o
-by-N output frame-based
signal.
For each output buffer, the block acquires the number of new input samples
specified by the difference between the
Buffer size (M
o
) and Buffer
overlap
(L) parameters. Each new input sample enters at the bottom of the
buffer, and is pushed upwards as later samples enter. The first row in the
output therefore corresponds to the earliest input sample. Because the block
can buffer a signal to a larger or smaller frame size, the number of samples
acquired from the input can be greater or less than the number of samples in
an individual input frame.
In general, the output frame period, T
fo
, is related to the input sample
period, T
si
, by
where M
o
is the Output buffer size (per channel), and L is the Buffer
overlap
.
As a result, the new output sample period, T
so
, is
Clearly, this is equal to the input sample period only when the
Buffer overlap
is zero. See “Converting Sample Rates and Frame Rates” on page 3-20 for more
information about rate conversions.
The following sections provide examples of buffering, and explore related
buffering issues:
•“Example: Buffering Sample-Based Signals without Overlap” on page 3-49
•“Overlapping Buffers” on page 3-50
•“Example: Buffering Sample-Based Signals with Overlap” on page 3-50
T
fo
M
o
L–()T
si
=
T
so
M
o
L–()T
si
M
o
-------------------------------=