CDB6420 Просмотр технического описания (PDF) - Cirrus Logic

Номер в каталоге

Компоненты Описание

Список матч

CDB6420

Full-Duplex Speakerphone Chip

Cirrus Logic 

CS6420

DESIGN CONSIDERATIONS

When designing the CS6420 into a system, it is im-

portant to keep several considerations in mind.

These concerns can be loosely grouped into three

categories: algorithmic considerations, circuit de-

sign considerations, and system design consider-

ations.

Algorithmic Considerations

The CS6420 facilitates full-duplex hands-free

communication via many algorithms running on

the Digital Signal Processor that is the core of the

CS6420. Among these are the algorithms that per-

form the adaptive filtering, the half-duplex switch-

ing, digital volume control, and supplementary

echo suppression.

Full-Duplex Mode

Full-duplex hands-free communication is achieved

through a technique called adaptive filtering. The

basic principle behind adaptive filtering is that the

acoustic path between speaker and microphone can

be modeled by a transfer function which can be dy-

namically determined by an adaptive digital filter.

This principle assumes good update control and

speech/tone detection algorithms to prevent the fil-

ter from mistraining.

Theory of Operation

Figure 8 illustrates how the adaptive filter can can-

cel echo and reduce loop gain. The echo path of the

system is between points B and C: the speaker to

A

B

F

Adaptive Filter

D-

E

Σ+

C

Figure 8. Simplified Acoustic Echo Canceller

Block Diagram

microphone coupling. A signal injected at A

(sometimes called a “training signal”) is sent both

to B, the input of the echo path, and to F, the input

of the adaptive filter. The signal at B is modified by

the transducers and the environment, and received

at point C (an “Echo”). Meanwhile, let us assume

for argument’s sake that the adaptive filter has ex-

actly the right transfer function to match the echo

path BC, and so the signal at point D is approxi-

mately equal to the signal at point C. After these are

subtracted by the summing element, all that is left

is the error signal at point E, which should be very

small.

If a person were to speak into the microphone at

point C, that signal would pass through the sum-

ming element unchanged because the adaptive fil-

ter had no comparable input to subtract out. In this

manner, the person at A and the person at C may si-

multaneously speak and A will not hear his own

echo.

In the real world, the echo path is not static. It will

change, for example, when people move in the

room, when someone moves the speaker or the mi-

crophone, or when someone drops a piece of paper

on top of the speaker. So, the filter needs to adapt

to modify its transfer function to match that of the

environment. It does so by measuring the error sig-

nal at point E and trying to minimize it. This signal

is fed back to the adaptive filter to measure perfor-

mance and how best to adapt, or train.

The trouble arises when the person at the near-end

(C) speaks: the error signal will be non-zero, but

the adaptive filter should not change. If it tries to

train to the near-end signal, the adaptive filter has

no way to reduce the error signal, because there is

no input to the filter, and therefore no output from

it. The adaptive filter would mistrain.

To prevent this mistraining, the echo canceller uses

double-talk detection algorithms to determine

when to update. These update control algorithms

22

DS205PP2

Share Link: