LG -P970 Service Manual page 56

3. TECHNICAL BRIEF
3.9.6 Audio function
Audio Processing on ARM Core
The ARM™1176 core is mainly in charge of "audio application" tasks, including audio playback, audio streaming
and audio recording if required. This implies the usage of codec algorithms such as MP3, WMA, AAC etc., which
must be implemented as decoder and/or encoder software on the ARM core. Decoded audio streams are
transmitted as PCM samples to the DSP core via a shared memory. Audio streams to be encoded are transmitted
by the DSP to the ARM core via a shared memory. The ARM can also generate ring tones and service tones to
notify an incoming call, an incoming message or any audio notification of a phone´s event (warning, error,
confirmation, battery low, etc.). Another activity on the audio side for the ARM is the streaming of compressed
audio data over the UART interface to the Bluetooth chip for A2DP usage (Advanced Audio Distribution Profile:
audio playback over Bluetooth). Via the same interface it communicates with the Bluetooth chip (this is not
restricted to audio activities). The ARM core is managing the configuration of the AFE by sending commands to
the DSP over shared memory to activate various configurations of the AFE. Note that the activities of the ARM
remain purely digital audio processing.
Audio Processing on DSP Core
The DSP core is in charge of "speech audio" and mixing tasks and its activities remain - as the ARM - purely
digital audio. The DSP can also execute some "audio application" tasks. It ensures the simultaneous encoding
and decoding of speech audio frames during telephone conversation. It uses different sets of vocoders as
specified by the applicable standards for the ongoing telephone conversation. On the audio out path (i.e.
towards the speaker) it mixes "audio application" received over shared memory with decoded speech audio. On
this path it can also mix some audio samples coming from the microphone to enable the feedback loop to the
speaker (Sidetone generation, as specified in 3GPP/ TS 43.050, chapter 3.10). On the audio in path (i.e. from the
microphone to the transmission of the encoded audio samples over the air) the DSP can also mix speech audio
samples with any other digital audio source and analogue FM radio as well. The mixing capabilities are
extremely broad as the cases mentioned above are only examples. These mixing capabilities also include the
ability to execute a sample rate conversion to bring all samples to the same rate before
mixing. The DSP also can execute some audio application tasks such as PCM, ADPCM, SB-ADPCM audio playback
and recording. Simple Tone (sine tones, DTMF, key click) can be generated as well.
The next activity of the DSP is transferring data to and from the hardware peripherals (AFE, I S0 and I S1).
2 2
Asmentioned in Figure 1 a fifo buffer is used to exchange stereo audio samples between the DSP and the AFE
on the microphone path and speaker path. When receiving a command from the ARM the DSP directly
configures the different modes of the AFE. The DSP also manages the I S interfaces I S0 and I S1. I S0 interface,
2 2 2 2
configured as master interface, is used to exchange speech audio samples at a rate of 8 or 16 kHz mono on both
directions with the Bluetooth chip in the case of a speech call using a Bluetooth headset. The I2S1 interface can
be used as slave or master interface, enabling the exchange of audio stereo samples at any common rate
between 8 and 48 kHz, potentially asymmetrically between RX and TX. This is to act as an audio bridge with an
external audio chip (application processor, audio processor,etc.). Note that the activities of the DSP - like the
ARM - remain purely digital audio processing. Refer to the document "system requirements specification" for
more details on the different audio application features supported by the chip.
LGE Internal Use Only - 56 - Copyright © 011 LG Electronics. Inc. All right reserved.
Only for training and service purposes