Embedding And De-Embedding Overview - Sonifex Redbox RB-VHDMA8 User Handbook Manual

1
INTRODUCTION
The de-embedding and embedding channel routing is controlled via the front panel
buttons and indicators. There is also LED indication for SDI input status and audio group
presence.
The unit can be remote controlled via Ethernet or serial port connections using the Sonifex
SCi software.
It has a triple rate SDI receiver with automatic input rate detection and equalisation along
with two re-clocked and individually buffered SDI outputs. It supports the full range of 3G,
HD and SD standards from NTSC and PAL up to 1080p 60Hz.
Each digital input is normally sample rate converted to 48kHz before embedding, so that
it is synchronous to the video input. Though sample rate conversion can be bypassed on a
per input basis allowing SMPTE-337M data to be embedded. All digital outputs are output
at 48kHz, synchronous to the video input.
There is independent level control for each digital input and output channel, which can be
adjusted from -24dB through to +24dB in 0.5dB steps.
The digital inputs and outputs are unbalanced AES specification, and can be connected
through either BNC or D-Type connectors. These I/O connections are paralleled, allowing
one type to be used per input or output.

Embedding and De-Embedding Overview

Audio embedding is the process of taking one or more audio signals, formatting them into
packets and then placing them into the ancillary or blanking space of a video signal. Audio
de-embedding is the process of retrieving this audio data from the video signal but not
necessarily removing it. These two processes are standardised by two SMPTE documents,
SMPTE-272M for SD-SDI and SMPTE-299M for HD-SDI and 3G-SDI.
Both of these standards allow for the embedding of 16 channels of 24-bit, 48kHz audio.
Which are divided down into 4 groups of 4 channels.
Each standard consists of a control packet and an audio data packet. The control packets
hold certain control information about the audio to aid in the de-embedding process or
any other receiving equipment. There is one control packet per group, these packets are
optional for SD. The control packet is normally placed in the horizontal blanking period
once per field or frame. The audio data packets contain the actual audio sample data.
There are considerable differences between SD and HD, the main ones being that SD data
packets contain only 20-bit data and are of variable length while the HD data packets
contain 24-bit data and are of fixed length. The HD packets also provide an error correction
and detection method. The SD standard consists of a further packet, the extended data
packet. These packets contain the 4 least significant bits of the audio samples to provide
full 24-bit audio.
3
Redbox Embedder/De-embedder User Handbook