AMBE-2000™ Vocoder Chip

User’s Manual Version 3.0

page 9

Confidential Proprietary, Subject to Change

2.2.2

Channel Interface Overview

The channel interface is meant to be flexible to allow for easy integration with the system under design. The basic hardware

unit of the interface is a serial port. The serial mode can run in passive or active modes. Simply stated, the control signal for

serial mode can be derived by the AMBE-2000™ chip or they can be derived externally.

Under normal operation, every 20msec the encoder outputs a frame of coded bits, and the decoder needs to be delivered a

frame of coded bits. There is some formatting of the data for both the encoder and the decoder. The primary purpose of the

formatting is to provide alignment information for the encoded bit stream. The data has two formats, Formatted and

Unformatted. Serial mode can run in either Formatted or Unformatted mode.

The Formatted and Unformatted modes are explained in full detail in Section 4, but essentially the two formats are trying to

achieve the same function, to provide positional information regarding the outgoing and incoming coded data streams. In

Formatted mode each 20msecs of output data from the encoder is preceded by a known structure. This structure also embeds

some status type flags, meant for local control purposes, within it. The only data from the Formatted format that is typically

sent across the transmission channel under design are the actual encoded bits at the desired rate.

In Formatted mode, it is the responsibility of the designed system to pass enough information along with the encoded bits such

that the Formatted format needed by the decoder can be reconstructed on the other side. This extra information, or overhead,

is going to be very specific to the system under design, but at a minimum needs to pass enough information to reliably

reconstruct the 20msec frame structure at the other end for the decoder.

In Unformatted mode the data coming out of the encoder can be thought of as a continuous stream of voice data with the

framing information embedded within the encoded bits. One advantage of this type of set-up is that the system does not have

to add any bandwidth for overhead to the channel. The disadvantage is that the decoder needs 10-12 incoming frames in order

to gain synchronization with the data stream before it can properly synthesize the speech waveform. Also, the Unformatted

mode only commits a single bit per frame to maintaining data alignment. In higher error rate channels the performance will be

improved by adding more bits per frame to the alignment information (which is more easily performed when using Formatted

mode)

Additional flexibility is given to the channel interface to the encoder and decoder by allowing the AMBE-2000™ Vocoder

Chip to run in Passive or Active modes. In Passive mode, data strobes are provided by an external source, while in Active

mode, data strobes are provided by the AMBE-2000™ Vocoder Chip. The serial interfaces can be run in Passive or Active

modes. See Section 4 for full details and timing for both Formatted and Unformatted data.

2.2.3

Speech and FEC Rate Selection Overview

The total coded bit rate is the sum of two components, the Speech Data and the Forward Error Correction (FEC) Data. The

addition of FEC data to the speech data allows the decoder to be able to correct a limited amount of errors within each frame

should they arrive corrupted. If the channel is expected to have more errors then more bits should be dedicated to FEC. At the

same time, voice quality will increase if the number of speech bits can be maximized.