Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Most of the signals generated are analog in nature. Hence these signals are
converted to digital form by the ADC. The DSP performs signal processing operations like
filtering, multiplication, transformation or amplification etc operations over these digital
signals. The digital output signal from the DSP is given to the DAC to generate analog
signal again.
Continuous values signal signal amplitude takes on all possible values on a finite or
infinite range
Discrete values signal. signal takes values from a finite set of possible values.
Periodic signal If x(n+N)= x(n) for all n where N is the fundamental period of the
signal. Else non-periodic signals.
Energy Signal Summation of magnitude squared values of x(n). The signal is called as
energy signal if its energy if finite. A signal is called power signal if its power is finite.
Ex: Energy of unit sample function is 1.
1
Discrete time frequencies = -1/2 to 1/2 cycles/sample or -∏ to +∏ rad/samples
Continuous time frequencies = - ∞ to +∞
5. Prove that discrete time signals are periodic only if frequency is rational.
What is the condition for periodicity of DT signal.
Causal If output of system depends only past and present inputs samples.
Non-causal If output of system also depends on future inputs.
2
Let X(n) = {2,5,2}. The signal x(n) can also be written as
X(n)= x(-1) δ(n+1) + x(0) δ(n) + x(1) δ(n-1).
14. Explain linear convolution. Will it applicable for Non-linear systems or Time
variant systems.
In linear convolution we decompose input signal into sum of elementary signal. Now
the elementary input signals are taken into account and individually given to the system.
Now using linearity property Whatever output response we get for decomposed input
signal, we simply add it & this will provide us total response of the system to any given
input signal.
Linear Convolution states that
y(n) = x(n) * h(n)
∞
y(n) = ∑ x (k) h(n – k )
k= -∞
3
Thus the frequency 50 Hz, 90 Hz , 130 Hz … are alias of the frequency 10 Hz at the
sampling rate of 40 samples/sec. To avoid aliasing sampling frequency should be selected
as per sampling theorem and pass the signal through pre-alias filter before sampling.
26. Explain the frequency relationships between continuous time and discrete
time signals.
Continuous time frequencies are given as Ω and F. while discrete time frequencies
are given as ω and f. Conversion relationships are given as ω = Ω Ts and f=FTs.
27. What is the use of correlation in DSP. How it is related with linear
convolution.
Correlations are nothing but establishing similarity between one set of data and
another. Correlation is closely related to convolution, because the correlation is essentially
convolution of two data sequences in which one of the sequences has been reversed.
Applications are in
1) Images processing for (in which different images are compared)
2) In radar and sonar systems for range and position finding in which transmitted and
reflected waveforms are compared.
3) Correlation is also used in detection and identifying signals in noise.
28. What is the relationship between difference equation and system function.
System function can be obtained by taking Z transform of the difference equation.
4
UNIT II, III Z TRANSFORM
1. What is Z transform and ROC. What is the usefulness of ROC. What are the
applications of Z Transform.
For analysis of continuous time LTI system Laplace transform is used. And for
analysis of discrete time LTI system Z transform is used. Z transform is mathematical tool
used for conversion of time domain into frequency domain (z domain) and is a function of
the complex valued variable Z. The z transform of a discrete time signal x(n) denoted by
X(z) and given as
∞
X(z) = ∑ x (n) z –n z-Transform.……(1)
n=-∞
Z transform is an infinite power series because summation index varies from -∞ to
∞. But it is useful for values of z for which sum is finite. The values of z for which f (z) is
finite and lie within the region called as “region of convergence (ROC).
ADVANTAGES OF Z TRANSFORM : For calculation of DFT, for analysis and synthesis of
digital filter, used for linear filtering, used for finding Linear convolution, cross-correlation
and auto-correlations of sequences.
ADVANTAGES OF ROC: ROC is going to decide whether system is stable or unstable, the
type of sequences causal or anti-causal & decides finite or infinite duration sequences.
2. How poles and zeros & ROC decides the causality and stability of system.
LSI system is stable if and only if the ROC the system function includes the unit
circle. i.e r < 1. Thus Poles inside unit circle gives stable system. Poles outside unit circle
gives unstable system. Poles on unit circle give marginally stable system.
LSI system is causal if and only if the ROC the system function is exterior to
the circle. i. e |z| > r.
4. Discuss the ROC of the signal & pole-zero plot of the signal.
Consider two cases
case 1: Infinite signal & case 2: Finite signal.
5
5. Discuss the nature of the signal.
6. ROC does not contains poles. Discuss the correctness of this statement.
7. Define pole and zero of the system. What poles and zeros are plotted with
respect to unit circle in z plane.
The frequency at which the magnitude of transfer function approaches infinity is
called pole and the frequency at which magnitude of transfer function becomes zero is
called zero. Unit circle is the frequency axis in z plane.
6
B0 BN-J
BN BJ
3. This process will continue until you obtain 2N-3 rows with last two having 3
elements. Y0,Y1,Y2
A digital filter with a system function H(z) is stable, if and only if it passes the following
conditions.
a. D(Z)|Z=1 >0
b. (-1)N D(Z)|Z=-1 >0
c. |b0|>|bN|, |C0|>|CN-1|
Z Transform Properties
Standard Z Transforms
7
UNIT IV: FT,DFT AND FFT
1. Why the frequency domain analysis is preferred over time domain analysis
in DSP.
Time domain analysis provides some information like amplitude at sampling instant but
does not convey frequency content & power, energy spectrum hence frequency domain
analysis is used. Magnitude and phase plot can be obtained from its FT and system
characteristic can be described well by using its frequency domain.
2. What is DTFT. Explain the nature of the spectrum of discrete time signal.
The discrete time Fourier transform of the signal is denoted as X(ω). It is also called
as analysis equation. It is given as
∞
X(ω) = ∑ x (n) e –jωn
n=-∞
Here ω is the frequency of discrete time signal and it takes all possible values between -∏
to ∏. Hence its Fourier transform is continuous.
Case 1: If x(n) is infinite or finite non-periodic sequence then its spectrum X(ω) is
continuous in nature.
Case 2: If x(n) is finite periodic sequence then its spectrum X(ω) will be discrete.
Inverse DTFT is also called synthesis equation. Here integration is used since X(ω) is the
continuous function of ω. Integration limits are -∏ to ∏. And the period of integration is
2∏.
8
Sr No Sequence DTFT
1 X*(n) X*(- ω)
2 X*(-n) X*(ω)
3 XR(n) Xe(ω)=1/2 [ X(ω) + X*(-ω)]
4 jXI(n) Xo(ω)=1/2 [ X(ω) - X*(-ω)]
5 Xe(n) XR(ω)
6 Xo(n) jXI(ω)
DTFT Properties:
DFT Properties:
9
5. Why DFT's are used in frequency domain analysis in place of DTFT.
FT is the continuous function of x(n) and the range of ω is from - ∏ to ∏ or 0 to 2∏.
while DFT is calculated only at discrete values of ω. Thus DFT is discrete in nature which is
sampling version of FT and thus mostly used in analysis of discrete signals.
For Discrete time signals x(n) , Fourier Transform is denoted as x(ω) & given by
∞
X(ω) = ∑ x (n) e –jωn
n=-∞
DFT is denoted by x(k) and given by (ω= 2 ∏ k/N)
N-1
X(k) = ∑ x (n) e –j2 ∏ kn / N
n=0
7. What are overlap save and add method. Why these methods are used.
When the input data sequence is long then it requires large time to get the output
sequence. Hence other techniques are used to filter long data sequences. Instead of finding
the output of complete input sequence, it is broken into small length sequences. The
output due to these small length sequences are computed fast. The outputs due to these
small length sequences are fitted one after another to get the final output response.
9. If input signal x(n) contains 4 samples. How many samples will be present
in its DFT. What will happen if it contains less than 4 samples.
10
If DFT is to be calculated at selected points only then, Goertzel algorithms are used.
Goertzel algorithms are used to calculated DFT as linear filtering operations and required
less number of calculations.
13. What mathematical tools are used to convert the signals from time domain
to frequency domain.
16. If two sequences are multiplied in time domain what will be effect on their
DFT's.
17. Circular Convolution can be obtained from linear convolution but vice-versa
is not possible. Discuss this statement with an example.
Table shows first column of memory address in decimal and second column as binary.
Third column indicates bit reverse values. As FFT is to be implemented on digital computer
simple integer division by 2 method is used for implementing bit reversal algorithms.
b WNr B= a - WNr b
From values a and b new values A and B are computed. Once A and B are
computed, there is no need to store a and b. Thus same memory locations can be used to
store A and B where a and b were stored hence called as In place computation. The
advantage of in place computation is that it reduces memory requirement. Thus for
computation of one butterfly, four memory locations are required for storing two complex
numbers A and B.
11
21. Can FFT Algorithms are applicable for the values of N which are not power
of 2. Example N=12.
Yes, In such cases sequence is padded with sufficient number of zeros such that the
value of N becomes the power of 2. Alternately (Another method)
If N=12, It can be divided into 3 sequence of 4 samples each. These sequences will be as
follows
First Sequence: x(0), x(3), x(6), x(9)
Second sequence: x(1), x(4), x(7), x(10)
Third sequence: x(2), x(5), x(8), x(11)
Now 4 point DFT's are calculated and then proceed further.
12
UNIT V DIGITAL FILTER
1. What is Sinc function.
Sinc pulse represents impulse response of ideal LPF while impulse train
represents ideal sampling function.
5. What are notch and Comb filters. What are its applications.
A notch filter is a filter that contains one or more deep notches or ideally perfect
nulls in its frequency response characteristic. Notch filters are useful in many applications
where specific frequency components must be eliminated. Example Instrumentation and
recording systems required that the power-line frequency 60Hz and its harmonics be
eliminated.
comb filters are similar to notch filters in which the nulls occur periodically across
the frequency band similar with periodically spaced teeth.
Frequency response characteristic of notch filter |H(ω)| is as shown
ωo ω1 ω
Impulse invariance method is generally used for designing low frequencies filter like LPF.
while for designing of LPF, HPF and almost all types of Band pass and band stop filters BZT
method is used.
12. Plot Mapping between analog and digital filter frequencies in BZT method.
-1
ω 2 tan (ΩT/2)
ΩT
14
13. What is frequency warping. Why it is used in filter design.
In BZT Frequency relationship is non-linear. Frequency warping or frequency
compression is due to non-linearity. Frequency warping means amplitude response of
digital filter is expanded at the lower frequencies and compressed at the higher frequencies
in comparison of the analog filter. But the main disadvantage of frequency warping is that
it does change the shape of the desired filter frequency response.
15. State the mapping between Z Plane and S plane in Impulse Invariance
method or Bilinear Transformation method.
1) Left side of s-plane is mapped inside the unit circle.
2) Right side of s-plane is mapped outside the unit circle.
3) jΩ axis is in s-plane is mapped on the unit circle.
Im[z] jΩ
1
Re(z) 3 σ
Z-Plane S-Plane
18. What are the various method used for FIR & IIR filter design
The various methods used for IIR Filer design are as follows
1. Approximation of derivatives
15
2. Impulse Invariance
3. Bilinear Transformation
The various method used for FIR Filer design are as follows
1. Windowing Method
2. DFT method
3. Frequency sampling Method. (IFT Method)
In Fourier series method, limits of summation index is -∞ to ∞. But filter must have
finite terms. Hence limit of summation index change to -Q to Q where Q is some finite
integer. But this type of truncation may result in poor convergence of the series. Abrupt
truncation of infinite series is equivalent to multiplying infinite series with rectangular
sequence. i.e at the point of discontinuity some oscillation may be observed in resultant
series.
Consider the example of LPF having desired frequency response Hd (ω) as shown in
figure. The oscillations or ringing takes place near band-edge of the filter. This oscillation
or ringing is generated because of side lobes in the frequency response W(ω) of the
window function. This oscillatory behavior is called "Gibbs Phenomenon".
22. For Speech processing or data transmission which type of filter are
preferred.
FIR filter always provides linear phase response. This specifies that the signals in the
pass band will suffer no dispersion Hence when the user wants no phase distortion, then
FIR filters are preferable over IIR. Phase distortion always degrade the system
16
performance. In various applications like speech processing, data transmission over long
distance FIR filters are more preferable due to this characteristic. Another reason is that
quantization noise can be made negligible in FIR filters.
24. Why FIR needs higher orders for similar magnitude response compared to
IIR filters.
Impulse response of ideal low pass filter is as shown in fig. In order to have finite
terms we will multiply this infinite series with rectangular window which will generate
desired frequency response. But some oscillation or ringing effect will be observed at the
point of truncation. This effect is known as Gibbs Phenomenon.
As M increases this side lobes becomes narrow and oscillatory behavior decreases.
As an example, the impulse response for a LPF is truncated with M=9,25 and an infinite
number of samples is as shown.
26. What are different window functions used for design of FIR filters.
Different types of windows functions are available which reduce ringing effect. These
are Triangular window, Blackman, Hamming window, Hanning Window and Kaiser window.
a. FIR filters designed using hamming window has reduced sidelobes compared to
rectangular window.
b. Blackman window has very small sidelobes but increased width of main lobe. In
Kaiser window has reduced side lobes and transition band is narrow and hence
mostly used.
17
27. What are the constraints to be imposed while designing filters from it pole
zero plot.
Filters can be designed from its pole zero plot. Following two constraints should be imposed
while designing the filters.
1. All poles should be placed inside the unit circle on order for the filter to be stable.
However zeros can be placed anywhere in the z plane. FIR filters are all zero filters hence
they are always stable. IIR filters are stable only when all poles of the filter are inside unit
circle.
2. All complex poles and zeros occur in complex conjugate pairs in order for the filter
coefficients to be real.
In the design of low pass filters, the poles should be placed near the unit circle at points
corresponding to low frequencies ( near ω=0)and zeros should be placed near or on unit
circle at points corresponding to high frequencies (near ω=∏). The opposite is true for high
pass filters.
28. Which window is better. Short duration window or long duration window.
Long Duration window. Because the length of window must be infinite in ideal case.
29. What are frequency transformation techniques. Why they are used.
Frequency transformation techniques are used to generate High pass filter, Bandpass and
bandstop filter from the lowpass filter system function.
Sr Type of transformation Transformation ( Replace s by)
No
ωhp
1 High Pass s
ωhp = Password edge frequency of HPF
(s2 + ωl ωh )
s (ωh - ωl )
2 Band Pass
ωh - higher band edge frequency
ωl - Lower band edge frequency
s (ωh - ωl)
s2+ ωh ωl
3 Band Stop
ωh - higher band edge frequency
ωl - Lower band edge frequency
18
UNIT VI: DSP PROCESSOR
1. What are the requirements of DSP processor. How It differs from general
Processor.
The most fundamental mathematical operation in DSP is sum of products also called as dot
of products.
Y(n)= h(0)*x(n) + h(1)*x(n-1) +………+ h(N-1)*x(n-N)
This operation is mostly used in digital filter designing, DFT, FFT and many other DSP
applications. A DSP is optimized to perform repetitive mathematical operations such as the
dot product. There are four basic requirements of DSP processor to optimize the
performance They are
1) Fast arithmetic
2) Fast Execution - Dual operand fetch
3) Fast data exchange
4) Circular buffering
5. What are the different functions used in MATLAB related with DSP.
Sr Function Application
No
1 conv(hn,xn) Linear Convolution of two sequences.
2 xcorr(x1n,x2n) Cross Correlation of two sequences.
3 xcorr(x) Auto correlation of sequence
3 fft(xn) DFT of x(n) using FFT algorithm
4 ifft(xn) IDFT of x(n) using FFT algorithm
5 Overlpsav (x,h,N) Implement Overlap save method to perform block
convolution.
21
5 zplane (b,a) Plot Pole zero plot.
6 freqz (b,a) Plot Magnitude phase plot.
7 freqs (b,a) Compute the frequency response of an analog filter.
8 bilinear(z,p,k,fs) Bilinear transformation
9 boxvar (M) Rectangular Window
10 hanning (M) Hanning Window
11 hamming (M) Hamming Window
12 kaiser(M) Kaiser Window
22