GA3216 Datasheet, PDF (9/14 Pages) Gennum Corporation – FOUNDATION-TM Digital Single or Dual Channel DSP Compression System

INTRODUCTION
The GA3216 hybrid comprises a configurable two-channel
compressor circuit with two methods of operation,
programmable mode and trimmer mode. It may be
configured as a one or two channel device with linear or
WDRC processing. Configuration data stored in non-volatile
memory defines hearing-aid parameters. This data needs to
be uploaded to the hybrid before the circuit becomes
functional. The GA3216 hybrid is programmed via the SDA
pin using industry-standard programming boxes.
Configuration data is generated by an ARK product
component library (DLL). Like the PARAGON™ Digital
products, the GA3216 is fully supported by Gennum's
software tools available from the Gennum ARK web site
http://ark.gennum.com.
During normal trimmer mode operation, a low-speed A/D
circuit monitors the positions of up to four manual trimmers
and a VC potentiometer. Trimmer position changes are
immediately interpreted and translated into new circuit
parameter values, which are then used to update the signal
path.
Gennum's Library Manager tool allows one to predefine
trimmer assignments for all common functions such as low-
cut, high-cut, notch and resonant-peak-shift filtering, AGC
parameters, wideband gain, and maximum power output
(MPO).
SIGNAL PATH
There are two main audio input signal paths. The first path
contains the Microphone and second path contains the
Telecoil input as selected by a programmable MUX. The
microphone input is intended as the main audio input for
single-microphone applications. The two audio inputs are
buffered, sampled and converted into digital form using an
A/D converter. The digital output is then converted into a
32kHz 20-bit digital audio signal. It is possible to save
current by reducing the sampling rate to 24kHz, which will
also reduce the systems bandwidth from 16kHz to 12kHz.
In Telecoil mode, gains are trimmed during the Cal/Config
process to compensate for microphone/Telecoil
mismatches. The Telecoil input may also be used as a
second microphone input. In this case the Telecoil
compensation would be disabled. This can be done via the
ARKonline wizard.
The wideband detector circuit output is routed to a band-
split filter that divides the signal into two frequency bands.
There is an option to configure the hybrid as a single
channel device that simply shuts down and bypasses the
band-split filter. The signal in each frequency band is
processed by an independent AGC loop. The gain in any
band is a function of the energy detected in that band or
the overall detected wide-band energy. The two frequency
bands are then summed back into a wide-band signal.
Further processing capabilities include a 1st to 3rd order
low cut filter before the band-split filter. After the two
channels are summed together there is a 1st or 2nd order
high cut filter, three EQ filters followed by volume control,
low-distortion AGCo, and two post-AGCo EQ filters.
All AGC loops use a feed-forward topology to ensure
system stability. Detectors in the AGCi path and AGCo path
are twin fast/slow detectors with independently adjustable
attack and release time constants. The squelch detector
(which resides in the AGCi path) is implemented as a single
fast/slow detector.
FUNCTIONAL BLOCK DESCRIPTIONS
A/D AND D/A CONVERTER
The system’s A/D converter is a 2nd-order sigma-delta
modulator, which operates at a 2.048MHz sample rate. The
system’s input is pre-conditioned with anti-alias filtering and
a programmable gain pre-amplifier. The analog output is
oversampled and modulated to produce a 1-bit pulse
density modulated (PDM) data stream. The digital PDM
data is then decimated down to pulse-code modulated
(PCM) digital words at the system’s sampling rate of 32kHz.
The D/A is comprised of a digital 3rd-order sigma-delta
modulator and an H-bridge. The modulator accepts PCM
audio data from the DSP path and converts it into a 32-
times oversampled, 1-bit PDM data stream, which is then
supplied to the H-bridge. The H-bridge is a specialized
CMOS output driver used to convert the 1-bit data stream
into a low-impedance, differential output voltage waveform
suitable for driving zero-biased hearing aid receivers.
CHANNEL PROCESSING
Figure 1 represents the I/O characteristic of independent
AGC channel processing. The I/O curve can be divided into
four main regions:
• low input level expansion (squelch) region
• low input level linear region
• compression region
• high input level linear region (return to linear)
0
High Level
-10
Gain
-20
-30
-40
Low Level
-50 ^Gain
-60
-70
-80
Compression
Ratio
Lower
Threshold
Squelch
Threshold
Upper
Threshold
-90
-100
-120 -110 -100 -90 -80 -70 -60 -50 -40 -30 -20
INPUT LEVEL (dBV)
Fig. 11 Independent Channel I/O Curve Flexibility
9 of 14
24501 - 3