20-07-2013, 03:00 PM
STUDY OF PCM VOICE CODING AND CODEC FREQUENCY RESPONSE
AIM:-
The aim of this experiment is to study the linearizized A-Law PCM cosing. The analog and
digital conversion as well as the reverse process and the filtering characteristics of the CODEC chip
145502 used in the OFC kit.
THEORY:-
Present techniques of voice communication use standards such as A-Law / U- law companded PCM
voice coding at 64kbits/sec. When the analog speech signal is converted to pulse code modulation, it
first is filtered using a low pass filter with a cut-off at about 3.4KHz. The analog signal is sampled at
8kHz as per the Nyquist Criteria. Each sample is quantized and coded into eight bits per sample.
Voice signal has a varying amplitude range that varies from one conversion to another. If the
quantization levels are uniformly spaced then it certainly creates problems such as.,
1. If the amplitude of the signal is small, quantization levels have to be closely spaced. This
gives proper resolution.
2. If the signal amplitudes are large then this fine resolution will result in increasing the no of
code bits.
3. Normally, unequal spacing of quantization levels are used.
The digital data output is in PCM form. The codec chip used exhibits both A-law and U-law
companding techniques.
PROCEDURE:-
1. Make connections as shown in the block diagram.
2. Voice communication can be carried out between the two kits using telephone handsets.
3. We can observe the effects of the voice at various test points.
4. Keep SW6 towards SINE IN position.
5. Feed a sinusoidal signal of 1kHz 2vP-P to SINE 1 & SINE 2 input terminals.
6. This gives an analog input to both the CODEC chips.
7. The reconstructed wave forms can be observed at the CODEC 1 RX and CODEC II RX pins.