SAA2013 Datasheet, PDF (9/32 Pages) NXP Semiconductors – Adaptive allocation and scaling for PASC coding in DCC systems

Philips Semiconductors
Adaptive allocation and scaling for PASC
coding in DCC systems
Preliminary specification
SAA2013
For each sub-band frame, SAA2013 calculates the
allocation and scale factor index information required by
the SAA2003. In order to synchronize the codec part of
SAA2003, SAA2013 frequently requests status
information from the codec. It monitors sample frequency,
emphasis information and stereo mode, and uses the
ready-to-receive bit of the codec to determine when to
transfer information.
Audio sample resolution section
The SAA2013 is designed for operation with audio input
sources of 14, 15, 16 or 18-bit resolution.
For optimum audio performance the bit allocation
algorithm of the SAA2013 can be varied to suit the bit
resolution of the audio source. This is done with the pins
RESOL0 and RESOL1 as shown in Table 1.
Decode mode
In decode the SAA2003 will transfer samples from FDAI to
FDAO with a delay of 480 FDWS periods. Settings and
status information can be sent to SAA2003 via SAA2013,
but the SAA2013 does not itself act on this information.
Transfer of this information is automatically synchronized
to the ready-to-receive bit of SAA2003 by SAA2013.
Table 1 Resolution set by pins RESOL0 and RESOL1.
RESOLUTION
16 bits
18 bits
14 bits
15 bits
RESOL0
0
0
1
1
RESOL1
0
1
0
1
Filtered data interface
The filtered data interface signals are given in Table 2.
Table 2 Filtered data interface signals.
PIN
FDWS
FDCL
FDAI
FDAO
FSYNC
INPUT/OUTPUT
input
input
input
output
input
FUNCTION
FREQUENCY
filtered data interface word select
f_s
filtered data interface bit clock
64f_s
filtered data input
−
filtered data output
−
filtered data sub-band synchronization −
The filtered data interface transfers sub-band filtered
samples between the stereo filter codec SAA2003 and
SAA2013. The interface is similar to a normal I²S interface,
consisting of clock (FDCL), data (FDAI/FDAO) and word
select lines (FDWS), except that the samples sent
represent signals divided into 32 sub-bands. One frame of
data consists of 12 samples from 32 sub-bands for both
left and right channels, i.e.: 768 audio samples. Each
audio sub-band sample is represented by a 24-bit two’s
complement number.
The order in which the samples are sent is shown in
Table 3.
For two channel mono, the order is the same, but with
Channel 1 samples in the place of left and Channel 2
samples in place of right.
Table 3 Order of samples.
SUB-BAND 0 0 1 1 2 2 ... 31 31
Channel
L R L R L R ... L R
Sample
0 0 0 0 0 0 ... 0 0
1 1 1 1 1 1 ... 1 1
2 2 2 2 2 2 ... 2 2
. . . . . . ... . .
. . . . . . ... . .
11 11 11 11 11 11 ... 11 11
The signal FSYNC is used between each PASC frame to
indicate the sending of samples for sub-band 0 (Fig.7).
May 1994
9