IBM39STB0210x
STB0210x Digital Set-Top Box Integrated Controllers
Advance
Architecture and Subsystem Information
Page 10 of 39
STB02_sds_0327.fm.01
March 27, 2000
Transport and descrambler features include:
Internal DVB (1.0 or 1.1) descrambler, including ltering and storage of eight control word pairs
Auxiliary output port for real-time data transfers:
- 8-bit mode at 1X, 1/2X, 1/3X, 1/4X and 1/8X of the system clock speed
Table section ltering:
- 64 separate 4-byte lter blocks with bit-level masking with full match/not match capability
- Multiple lters can be linked to extend ltering depth in 4-byte increments
- Multiple lters per PID
- Filters program-specic information (PSI), service information (SI), private tables
- Handles multiple sections per packet and sections that span packets
- Optional CRC checking of section data
Selective routing of some or all packet data to system memory:
- Based on 32 separate queues (one per PID)
- Routing entire packets, payloads, adaptation elds, table sections (after ltering) and private data
Direct transfer of audio / video (PES) data to decoders
Simplied channel changes, time-base changes and error agging / concealment through direct commu-
nication with decoders
MPEG-2 Audio Decoder
The Audio Decoder receives and decodes either ES (Elementary Stream) or PES (Packetized Elementary
Stream) audio data. The audio compute engine is a generic DSP processor that decodes MPEG, or 16-, 18-
or 20-bit unformatted Pulse Code Modulation (PCM) audio data via individual software programs.
The host processor downloads each program load to the Audio Decoder following initialization. The Audio
Decoder generates up to two channels of decoded PCM for MPEG and PCM audio playback output. It pro-
vides 2-channel MPEG audio to two channels output. Unpacketized PCM (UPCM) plays back at sampling fre-
quencies of 16 kHz, 22.05 kHz, 24 kHz, 32 kHz, 44.1 kHz, and 48 kHz, along with quantization sample width
selections of 16-, 18-, or 20-bit input and 16 or 20-bit output.
The Audio Decoder:
Decodes MPEG-1 and MPEG-2 audio, Layers I and II and 2-channel output, including single channel,
stereo, joint stereo, and dual channel modes.
Performs MPEG-1 and MPEG-2 PES audio parsing, and also accepts audio elementary streams. Parses
and stores ancillary data into external memory for later use by the host processor.
Supports 16-kHz, 22.05-kHz, 24-kHz, 32-kHz, 44.1-kHz, and 48-kHz audio sampling frequencies.
Supports audio/video synchronization through PTS/STC comparison with each audio frame.
Supports an encoded audio bit rate up to 640 Kbps. This bit rate only pertains to encoded bitstream data.
Includes Audio Clip Mode for PES, ES, and PCM formats with byte address granularity and 2MB maxi-
mum per clip buffer.
Supports expandable rate buffer size selectable from 4K to 64K (in 4K increments).
Uses a re-locatable rate buffer region, with a programmable base register (128-byte location granularity).
Has a re-locatable PTS Value and Ancillary data region, using a programmable base register with 128-
byte location granularity.