Handbook for Sound Engineers

1466 Chapter 39

The state of the beginning of the preamble must
always be opposite that of the second state of the parity
bit that ends the subframe before it.

You will note that the two versions of each preamble
are simply polarity reversed versions of each other.

In practice, due to the nature of the positive parity
used for the bit before the preamble, and the biphase
coding, only one version of each preamble will ever be
transmitted. However, to preserve the insensitivity to
polarity inversions, AES3 receivers must be able to
accept either version of each preamble.

Like biphase-mark coding, the preambles are dc free
and allow for clock recovery while differing from valid
biphase-mark coding at least twice.

The clock rate shown above is at twice the source bit
rate. Note that the second state of the parity bit is always
zero, and therefore the preamble will always start with a
transition from zero to one. Also note that in this
preamble, as in all possible preambles, there are at least
two places where there is no transition at a bit boundary
thus violating the rules for biphase-mark coding and
providing positive identification of the preamble.

39.2.5 Channel Status Format

Each audio channel has its own channel status bit. The
data carried by that bit is associated with its own audio
channel. There is no requirement that the data for each
channel be identical, although it could be, Fig. 39-7.

The sequence of 192 channel status bits in a given
block is treated as 24 bytes of data, as shown in Table
39-1.

The sequence of channel status bits for each channel

starts in the frame with Preamble Z.

39.3 AES3 Implementation

39.3.1 AES3 Transmitters

As a minimum, an AES3 transmitter must encode and

transmit the audio words, validity bit, user bit, parity

bit, the three preambles, and a minimum version of the

channel status.

The minimum version of channel status will have

byte 0 bit 0 set to one to specify this is a “professional

use of channel status block” and all the other bytes set

to their default values.

Some AES3 receiving devices might have problems

with such a minimum version of channel status for two

reasons. First, many receivers expect to see a properly

encoded CRC in byte 23, and will therefore show a

CRC error when receiving the default 0’s instead of a

CRC. Second, some receivers might expect to see the

sampling frequency in byte 0 bits 6–7, and not have

provision for manual override or auto set of the

sampling frequency.

Even if some addition information is included in the

channel status beyond what is listed as a minimum

above, unless all the information considered standard

below is included, the interface must still only be called

a minimum implementation of AES3.

A standard implementation will include everything

specified as minimum above plus will encode and

transmit all the information in bytes 0, 1, 2, and 23 of

the channel status.

An enhanced implementation provides additional

capabilities beyond the standard implementation.

All transmitters must be documented as to which of

the channel status capabilities they support.

39.3.2 AES3 Receivers

All receivers must document the level of implementa-

tion provided and the actions that will be taken by the

receiving device based on the information received.

39.3.3 Electrical Interface

AES3 uses a balanced 110: electrical interface based

on the International Telegraph and Telephone Consulta-

tive Committee (CCITT) Recommendation V.11. It is

Figure 39-6. AES3 Preamble X (11100010). The time
between clock pulses is called the unit interval (UI).

Preceding

State

01

Channel Coding

“X” 11100010 00011101 Subframe 1

“Y” 11100100 00011011 Subframe 2

“Z” 11101000 00010111 S u b f r a m e 1 a n d b l o c k

start

1 1 1 0 0 0 1 0

Parity LSB

Lack of transition

at bit boundary

Clock