how to classify signals?
discreet time/continous time, periodic/nonperiodic
what is the use of random signals?
for testing systems dynamic response statistically for very small amplitudes and time durations.
how to classify systems?
linear, stable, time invariant
what's the difference between correlation & convolution? how to obtain convolution from correlation operator *?
conv(h,x) = h*(-x)
what are the characteristics of a transient response of a system?
decay time, rise time, peak time, max overshoot, settling time
whats sampling thereom? who proposed it? whats importance?
Why is the need of FFT ?
What is the need of Digital Signal Processing?
what is the convolution?
What is aliasing and how do we prevent it?
Aliasing is frequency shifting of content of input signal above half the sample frequency. We avoid it by using antialiasing filters to limit signal content below 1/2 sample frequency and/or by sampling at high enough frequency to avoid antialiasing.
What is the application fo Cross correlation and Auto Correlation?
what do u mean by spectral resolution?
i think its separetion of two signal in frequency domain.
how do u reduce spectral leakage?
By incresing the window lenth in time domain we can reduce spectral leakage.
What is the concept of stability of an LTI system? How to check if a given system is stable?
Determine the roots of the system. If all roots have real part <> 0, the system is unstable.
What is the importance of time frequency resolution? Can this be explained with respect to some application eg when we need to determine the stationary and dynamic characteristic of a signal, how do transform with different time frequency resolution, affects it.
Time and frequency are two important aspects in which time and frequency are the inverse of each other.Hence if u have a small frequency it is spread out in large time and vice versa. Just u can analyze a time frequency plot.
So when we try to analyze non stationary signals these values continuously change and the properties of FFT do not hold for non stationary signals for which u need to go to either Short term Fourier trnasform or the wavelet analysis.
Each has it's own adv and disadv.
Tell some thing about Interpolation and decimation?
Intepolation basically filling missing samples,
decimation basically reducing the sample rate.
Explain using convolution the effects of taking an FFT of a sampe with no windowing (rectangular window).
The signal is multiplied by a rectangular window in the time domain. This corresponds to convolution by a sync function in the frequency domain. The wide center lobe of the sync function spreads frequencies and the side lobes shift frequencies.
What is Gibbs phenomenon?
The Gibbs phenomenon in time is the result of truncating its
corresponding discrete Fourier Transform, or, Fourier Series. The Gibbs
phenomenon in frequency is the result of truncating its sampled time series.
If a have two vectors, how will i check the orthogonality of those vectors?
Why do we need I&Q signals?
what care should be taken when we doing camulative decimation?
Transforms
In signal processing, why we are much more interested in orthogonal transform?
What are the pros and cons of Discrete Cosine Transform?"
What are basis vectors in a transofrm?
how can you compute fourier transform form Z-transform ?
z =r*e(-iw) when r=1 Z-transform converges to discrete Fourier tranform.
Why is FFT faster than DFT? what is the actual concept behind this?
FFT is an algorithm to implement DFT , FFT is not at all a tranform,
if u implement Directly it take (N^2) complex Multiplications and (N^2 -N) complex additions , where as FFT will Nlog2(N) complex Mul & additions
what is the difference between DFT and DTFT?
DFT = spectrum of DFT also discrete in nature,where as spectrum of DTFT is continuous.
why we use DCT extensilvely in compression?
DCT has excellent energy compaction property, for highly correlated input it give excellent energy compaction similar to KL transform.
why after DCt we use a zig zag manner for run length coding?
to convert to 2d to 1d data for run lenth coding, zig-zag is prefered because it scan the 2d data from low frequency to high frequency coef.
FFT is in complex domain, how to use it in real life signals optimally?
use two diiferent buffers for real & complex
why we use Forrier transform in DSP
how to find energy density (enrgy per unit bandwidth) from fourier transform?
| F(jw)| pow 2
why do we use laplace transform?
can u compare laplace & fourier transforms? any disadvantages for fourier?
in a strict sense, fourier is not applicable for unit step, unit ramp and sinusoidal functions. this is overcome in laplace.what are advantages of laplace transform?
simplifies operations, transforms frequently occuring exponential function to simpler easier to handle algebraic functions.
what's the fourier transform of a auto correlation for a periodic waveform ?
power spectral densityFilters
what is use of windowing in digital filters?
How toget a magnitude response of a quantized filter?
Any noise estimation has to be done??
What is the difference between equiripple filter and FIR filter?
whats basic difference b/w winer filter and kalman filter and lms filter ?
What is the advantage of a Direct form II FIR over fom I?
Why IIR filters doesnt have Linear phase?.
Liner phase IIR filter cannot physically realizable , its unstable.
Can IIR filters be Linear phase? how to make it linear Phase?
Idealy Physically realizable doesnt have linear phase, but we can implement IIR filter with linear phase response in passband. (refer Bessel series Approx)
What is the advantage of a Direct form II FIR over fom I?
i think Direct form -2 take less memory (not sure)
What is the basic difference blw FIR and IIR filters?
basic difference is FIR has liner phase response,where has IIR filter has nonlinear phase response.
what is the special about minimum phase filter ?
minimum phase system has both poles and zeors reside in side unit circle, these are basically used for
compensating channel impairments.
differences b/w butterworth, chebyshev, elliptical filter and advantages/disadvantages of each?
butter worth -monotonic response in pass band and stop band.
chebyshev-1 ripple in pass band, and monotonic in stop band.
chebyshev-2 monotonic in passband ripple in stop band.
elliptic- ripple response both in passband &stop band.
for given set of parameters we can achive minimum transistion band width with elliptic filter , other way
we can implement with fewer number of coef than other methods
What is Interpolation and decimation filters and why we need it?
interpolation for reconstruction and for sample rate up conversion.
deciamtion filter sample rate down conversion.
what is the simplest high pass filter ? write the equation?
Hp= 1- Lp