Scribd
Upload
222 views

Differential Pulse Code Modulation

DPCM and ADPCM are techniques for encoding signals that take advantage of correlation between samples. DPCM encodes the prediction error rather than the raw samples to remove redundancy. It predicts the next sample and encodes the difference between actual and predicted. ADPCM further improves on this by using adaptive quantization and prediction tailored to the current signal characteristics. This allows it to code speech at 32kbps compared to standard PCM's 64kbps, using fewer bits per sample and better following non-stationary speech signals. Both are susceptible to errors when prediction fails but ADPCM minimizes this through its adaptations.

Uploaded by

Narasimhareddy Mmk
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
222 views

Differential Pulse Code Modulation

DPCM and ADPCM are techniques for encoding signals that take advantage of correlation between samples. DPCM encodes the prediction error rather than the raw samples to remove redundancy. It predicts the next sample and encodes the difference between actual and predicted. ADPCM further improves on this by using adaptive quantization and prediction tailored to the current signal characteristics. This allows it to code speech at 32kbps compared to standard PCM's 64kbps, using fewer bits per sample and better following non-stationary speech signals. Both are susceptible to errors when prediction fails but ADPCM minimizes this through its adaptations.

Uploaded by

Narasimhareddy Mmk
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 12

DPCM

Differential Pulse Code modulation


DPCM
Voice and Video signals when sampled at a rate above nyquist rate results in high degree of
correlation between samples.
Because of which variance is smaller in difference signal than the variance of the signal itself.
 In PCM system, because of correlation between the samples, the resulting encoded signal
contains redundant information. The basic idea of DPCM is to generate an efficient coded signal
by removing the redundancy before encoding.
DPCM includes delta modulation as a special case.
One bit qunatizer
Zero prediction order ( single delay element)

DPCM, like DM, is subject to slope-overload distortion whenever the input signal changes too
rapidly for the prediction filter to track it. Also, like PCM, DPCM suffers from quantization noise.
Schematics of DPCM: Tx[1]
DPCM: working
Let us consider, m[n] is the sample at time instant n, and 𝑚 [n] be the prediction to m[n] at time
instant n.
Then the error : e[n] = m[n] – 𝑚 [n]
Then the quantizer output is given as
eq[n] = e[n] + q[n]
The prediction filter input is given as
mq[n] = 𝑚 [n] + eq[n]
mq[n] = 𝑚 [n] + e[n] + q[n]
mq[n] = m[n] + q[n]
mq[n] = Quantized version of sample m[n]. Irrespective of prediction filter
properties the quantized sample differs from the original sample by a
quantization error.
If the prediction is good, the variance of the prediction error will be smaller than
the variance of m[n], so that a quantizer with a given number of levels can be
adjusted to produce a quantization error with a smaller variance than would be
possible if the input sample m[n] were quantized directly as in a standard PCM
system.,
Schematics of DPCM: Rx [1]
ADPCM
In PCM, speech coding standard rate is given as 64kbps, which demands high
channel bandwidth.
In some applications, speech coding at low bit rates along with acceptable fidelity
is required. For Coding speech at low bit rates, a waveform coder of prescribed
configuration is optimized by exploiting both statistical characterization of
speech waveforms and properties of hearing.
The two aims in mind are:
◦ To remove redundancies from the speech signal as far as possible.
◦ To assign the available bits to code the non redundant parts of speech signal in a
perceptually efficient manner.
ADPCM
The main advantage of ADPCM is that the coding of speech reduces to 32kbps
through the combined use of adaptive quantization and adaptive prediction.
In ADPCM the number of bits are reduced to 4 bits instead of eight as used in
standard PCM.
The term adaptive means being responsive to changing level and spectrum of the
input speech signal.
The variations in speakers and speech material makes adaptive quantization and
adaptive prediction necessary to achieve best performance over a wide range of
speakers and speaking situations.
Adaptive quantizer: quantizer that operates with a time varying step size.
AQB
Where, is an estimate of the standard deviation.
The above equation can be implemented in two ways:
◦ Adapative quantization with forward estimation: Uses unquantized sample to
derive estimates
◦ Adaptive quantization with backward estimation (more preferred): Uses
quantized output to derive estimates and there by calculation if step size.
AQB schematic [1]
Schematics of ADPCM: Tx[1]
References:
[1] “Communication Systems”, 4th Edition by Simon Haykin, Wiley Publications.

[2] “Analog and Digital Communications: Theory and Lab Work”, 1 st Edition by
Abhay Gandhi, Cengage Publications.

[3] Hashima, Sherief. (2018). Digital Communication Systems.

You might also like