4 call testing and coefficient optimization, 1 far-end single-talk counting, 1 subtest a, ec convergence test – Cirrus Logic AN168 User Manual

Page 23: An168

Advertising
background image

AN168

AN168REV2

23

the two RVol values in dB.

1.5.4

Call Testing and Coefficient Optimi-
zation

The following tests are useful for optimizing the
CS6422 register settings. These tests should be
performed in a car, not a lab. They can be
performed in a lab, but the resulting coefficients
and settings will not be optimal for actual car use.

First, a note on safety. One of the compelling forces
driving the use of hands-free kits in cars is added
safety. Many governments have realized the
dangers inherent when people drive and talk on the
phone at the same time and have mandated the use
of hands-free accessories when cell phones are
used in cars. The studies do not mention the
dangers involved in driving while fine-tuning
hands-free system parameters.

Please, be careful.

Optimal car testing requires a minimum of three
people: a driver, a passenger, and a far-end talker.
The driver’s responsibility is solely to drive the car
and to avoid accidents. The passenger and the
far-end talker perform the call testing and
parameter optimization.

Once the test hardware is installed in the car, set the
Acoustic Coupling to a value of -9 dB using the
techniques discussed earlier in this note. Then
begin testing.

1.5.4.1 Far-end single-talk counting

The first test is the Far-End Single-Talk Counting
test. In this test, the far-end talker speaks, usually
incremental counting, and the near-end talker
remains silent.

This test has several sub-tests which are used to
configure specific parameters.

1.5.4.1.1 Subtest A, EC Convergence Test

The EC Convergence Test allows the far-end
listener to hear the echo canceler train, and retrain

on changes in path. In this test, the CS6422 is
loaded with the starting register configuration set
listed earlier in this note, which is repeated here for
convenience:

Configure the CS6422 from reset with the
following:

1) Mic set to '1' or '0', depending on whether the

internal mic preamp is used or not

2) GB = 0.75 dB/ms

3) RVol = +18 dB (this is the default setting; RVol

should be set between +6 dB and +30 dB)

4) Taps = 55.5 ms

5) TVol = 0 dB (this is the default setting; close to

0 dB is better; TVol should be between 0 dB
and +12 dB)

6) NseRmp = 12 dB/s

7) HDly = 150 ms

8) IdlTx = enabled

9) TSAtt = 24 dB

10) PCSen = low

11) TSThd = 12 dB

12) TSBias = 18 dB (default setting)

13) AErle = 18 dB

14) AFNse = -42 dB

15) TGain = 0 dB (can be 0 dB, +6 dB, +9.5 dB, or

+12 dB, depending on mic preamp require-
ments)

16) NECD = ‘1’ (should be '0' if a network sidetone

is present)

17) ASdt = -18 dB

The register configuration which implements the
above is:

reg 0: 1400 (9400 if internal mic preamp is
used)

reg 1: 0a22

Advertising