iDocSlide.Com

Free Online Documents. Like!

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

Share

Description

3UHIDEULFDWHGEXLOGLQJEORFNRIPDQ\$1'25JDWHV
½ DFWXDOO\125RU1$1'
½ SHUVRQDOL]HGE\PDNLQJRUEUHDNLQJFRQQHFWLRQVDPRQJWKHJDWHV
½ SURJUDPPDEOHDUUD\EORFNGLDJUDPIRUVXPRISURGXFWV

Tags

Transcript

AIM:
To study the adaptive delta modulation.
APPARATUS
:
1. Function generator 2. Clock generator 3.
Transmitter
(Differential amplifier, Comparator, Bistable circuit, Unipolar Bipolar converter, Integrator, Adaptive control circuit) 4.
Receiver
(Bistable circuit, Unipolar bipolar convertor, Adaptive control circuit, Integrator, Low pass filter)
THEORY:
Adaptive delta modulation
In delta modulation there are quantization errors due to slope overload and granular noise. To overcome these distortions, we use adaptive delta modulation in which step size is made adaptive to variations in the input signal. Adaptive delta modulation (ADM) or continuously variable slope delta modulation (CVSD) is a modulation of DM in which the step size is not fixed.
1.
In ADM, the processor has an accumulator and at each active edge of clock waveform, generate a step.
2.
In ADM step size is not fixed. Here to generate a step in response to the INPUT a clock is generated having clock edge(k-1).
3.
If the direction of step is same at the clock edge (k-1), then the processor increase the magnitude of step by amount. If the direction of step is opposite at clock edge (k-1) then the processor decreases the magnitude of step by amount.
4.
The output signal So (t) is called e(k), where e(k) represent an error signal. e(k)=+1 if m(t)>m^(t) immediately before the kth edge e(k)=-1 if m(t)<m^(t) immediately before the kth edge the step size S(k)=|S(k-1)|e(k) + Soe(k-1).
ADAPTIVE DELTA MODULATOR: ADAPTIVE DELTA DEMODULATOR:
THEORY OF CIRCUIT:
1.
The control circuit compares the present data bit from D flip flop with the previous two data bits. 2.
Its output to the counter is high when three bits are identical; the control circuit output goes low, thus setting the counter advance with every clock cycle. Each time counter is incremented from 00 to 11 is reached where it remains in that state. 3.
When slope overloading is not occurring, the integrator output always hunt above and below the analog input even after it is caught up with it. 4.
The output from a D flip flop is a constant change 1 to 0 at Tx clock edge. 5.
The changing input to the control circuit ensures that output to the counter is high and hence the counter is resettled at every clock cycle. 6.
Thus, the control word from counter is always 00 forcing the integrator gain at its lowest value thereby reducing quantization noise. 7.
Now a fast changing analog signal appears at the input of the modulator such that slope overloading occurs. The integrator output no longer follows the analog signal but it spends its time trying to catch up the analog signal. 8.
The output from a D flip flop is a constant change from 1 to 0 for 3 consecutive times. 9.
As soon as the third continuous 1/0 is sensed by the control circuit its output is low. 10.
The counter now advances to 01 doubling the integrator gain. This increases the ramping rate of integrator and it is able to catch the analog signal faster. 11.
In the next clock cycle if the same situation continuous, the counter advances to 10 forcing the integrator gain to 4 times its standard value. 12.
This situation continuous till the counter advances to 11 where it remains locked until control logic does not detect a change in the bit level at its input. 13.
As soon as the control circuit detects a change in bit level, its output goes high, thus resetting the counter and normalizing the integrator gain. 14.
The demodulator of the in here spikes. The output from integrator is passed to a LPF to smooth out the waveforms.
ADM ALGORITHMS SONG ALGORITHM: -
In the year of 1971, C. L. Song et al. [7] proposed a nice algorithm for step size adaptation by which the step size of the predicted waveform
(
delta
m
(
t
))
can nicely be adapted. According to song algorithm, positive slope of the prediction with respect to time results in the next prediction equal to the previous step size added with the step size at 0th prediction
(delta
0
)
, similarly, for negative slope of the prediction with respect to time results in the next prediction equal to the previous step size minus the step size at 0th prediction
(delta
0
)
. If the
n
th prediction sample is
delta(
n
)
, the algorithm prescribes the adaptation as shown in the following two equations.
….(1.1)
…..(1.2)
This addition and subtraction is called as
accumulation
in other terms. To combine the Eqs (1.1) and (1.2) let us assume another parameter called as discrepancy parameter
(
dn)
as
dn
=+
1
if m
(
t
)
>
_
m
(
t
)
−
1
if m
(
t
)
<
_
m
(
t
) ... (1.3) Now, combining Eqs. (1.1) and (1.2), the nth sample can be calculated using song algorithm of adaptation as
…..(1.4)
The song algorithm can now be used for digitization of any type of analog signal shape. Let us take a typical step size analog signal for prediction through accumulation. The ADM waveform through song algorithm is shown below. in the Fig., the ADM predicted waveform is showing undue oscillation for typically step like analog signal. For low slope and high slope signals of other types can be tracked more efficiently by this method of adaptation. The slope overload problem is solved totally by the song algorithm. The undue oscillation can be viewed as a special type of granular noise.

Related Documents

We Need Your Support

Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...Sign Now!

We are very appreciated for your Prompt Action!

x