AES3-1985, EBU Tech. 3250-E, CCIR Rec. 647, SP/DIF, IEC 958, EIA CP340 and EIA DAT are majors digital audio standard for serial interface transfer. These standards describe a unidirectional, self clocking, two channel standard based on a single serial data signal. The format contains audio samples up to 24 bit in lenght and non audio data including channel status, user data, parity and sample validity.
The differencesbetween these standards lie in electrical levels, connectors, and the use of the channel status bits.
One serial frame of information is transferred each sample period, each consisting of 64 bits, and comprising two subframes, A and B, one per each audio channel, as shown in fig. 1. A group of 192 consecutive frames forms a block.
Fig. 1 - AES/EBUC/CP340 Digital Audio Frame
Preamble patterns are used to synchronize and identify subframes and blocks. There are tree types of preambles (fig. 2), named X, Y and Z or 1,2 and 3 in AES specification and B,M and W in CP340 specification. The Z(1,B) preamble signifies the channel A subframe and also the beginning of a block. The X(2,M) preamble signifies the channel A subframe, any frame within the block other then the first. The Y(3,W) preable signifies the channel B subframe anywhere within the block.
Fig. 2 - Preambles for AES3, EBU Tech. 3250-E, and CP340
The channel A audio samples are referred to as A0-A23 and the channel B audio sample are referred to as B0-B23. A23 is the MSB (Most Significant Bit) of the channel A audio sample and B23 is the MSB of the channel B audio sample for all serial interface modes. If 16 bit audio samples are used, A8 and B8 are the LSBs (Last Significant Bits). The auxiliary data bits may be used to extend to audio samples to 24 bits or may be used for other purposes.
The validity (V) bit indicates if the audio sample data bits are valid (error free) and is commonly used to mute automatically D/A cnverters if the data is invalid. The bit is defined as a logic zero (0) if the audio sample is valid, and a logic one (1) if the sample is defective. The default value is a logic zero.
Each block provides 192 channel status bits (C). These forms 24 bytes of channel status data. These bits may be used according to the implemented standard, as explaned in each specific standard sections to be discussed later.
The user bit (U) is free for user data of any type or format desired. The CD format implements subcode blocks of 1176 user bits, organized as 98 subcoding symbols, with twelve user bits per subcode symbol. The DAT format uses user bits for SYNC, Stard-ID and Shortening-ID information AES18-1992 specifies an optional usage of the user bits as well.
The parity bit (P) implements even-parity transmission error detection. Since the preambles have even parity as an explicit property the parity bit can be taken as applying to all the data in a subframe.
[ Index | Next Paragraph ]