Waveform Quantization and Compression - Digital Signal Processing: Fundamentals and Applications - page 517

with the following constrains:

b 0 ðnÞ¼b 1 ðnÞ¼b 2 ðnÞ¼b 3 ðnÞ¼b 4 ðnÞ¼b 5 ðnÞ¼ 0f t r ðnÞ¼ 1 (11.25)

( 1

a 2 ðnÞh 0 : 71875

t d ðnÞ¼ (11.26)

0 otherwise

( 1

d q n > 24 , 2 y l

a 2 ðnÞh 0 : 71875 and

t r ðnÞ¼ (11.27)

0 otherwise

t d ðnÞ is the indicator that detects a partial band signal (tone). If a tone is detected ( t d ðnÞ¼ 1), Equation
(11.13) is invoked to drive the quantizer into the fast mode of adaptation. t r ðnÞ is the indicator for
a transition from a partial band signal. If it is detected ( t r ðnÞ¼ 1), setting the predictor coefficients to be
zero as shown in Equations (11.23) and (11.25) will force the quantizer into the fast mode of adaptation.

Simulation Example
To illustrate performance, we apply the ADPCM encoder to the speech data used in Section 11.1 and
then operate the ADPCM decoder to recover the speech signal. As described, the ADPCM uses 4 bits
to encode each speech sample. The MATLAB implementations for the encoder and decoder are listed
in Programs 11.11 to 11.13 in Section 11.7 . Figure 11.13 plots the original speech samples, decoded

5000

0

-5000

0 200 400 600 800 1000 1200 1400 1600 1800 2000

5000

0

-5000

0 200 400 600 800 1000 1200 1400 1600 1800 2000

1000

0

-1000

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Sample number

FIGURE 11.13

Original speech, quantized speech, and quantization error using ADPCM.