DIGITAL SIGNAL

PROCESSING
Bushra Bashir Chaoudhry
Last week we left over here...
Analog to Digital Representation
■ A signal changes itself subjected to some independent variable.
■ For instance consider a sinusoid signal as function of time i.e. .
■ can be represented in digital domain by doing two things
I. Sampling
II. Quantization.
Sampling
■ Obtaining the signal values from the continuous signal at regular time-intervals is
known as “sampling”.
■ The sampling interval is denoted as . Its reciprocal is sampling frequency .
■ The result of sampling process is a sequence of numbers, which can be represented as
where is the index of the this discrete sequence.
■ Sampling just extracts the signal values at all integer multiples of such that our discrete
time sequence becomes:

■ At this point after sampling we have a discrete time signal and not digital signal as yet .
WHY ?
Fig 2.1: Sampling a sinusoid: we measure
the signal values at regular intervals
Interpolation
■ The continuous signal x(t) is defined for all values of t, our discrete-time signal
is only defined for times which are integer multiples of .
■ To reconstruct the signal we need to guess what was in-between these sampled
values, this guess work is done by “interpolation”.
■ Interpolation is the process of ’guessing’ signal values at arbitrary instants of
time, which fall - in general - in between the actual samples.
■ There are different ways to do interpolation.
1. Piecewise constant interpolation: Crudest method, generates a stair-step
shape signal
2. Linear interpolation: To reconstruct a signal value we simply connect the
values at our sampling instants with straight lines. Better than piecewise
constant interpolation
 Interpolation (Cont’d)
3. More sophisticated interpolation methods such as higher order polynomial
interpolation (quadratic, cubic, quartic, pentic, etc.), (windowed) sinc- interpolation exist
as well - they will yield smoother and more faithful reconstructions.

a) b)
Fig 2.2: Reconstruction of signals using a) Piecewise constant interpolation
b) Linear interpolation
Assignment no. 1 [10]
■ What is interpolation? [1]
■ What is
1. piecewise constant interpolation
2. Linear interpolation
write details (5 lines each!) about each method. [2]
■ Demonstrate the two types with the help of numerical examples.[3+3].

■ P.S: Please keep the examples short, and complete the assignment in ‘two A4 pages’.
■ For P.S you’ll get an extra mark! [1]
Sampling Theorem
Sampling
■ CT signal DT Signal
Interpolation

■ However - our reconstructed (interpolated) continuous time signal is by no means

guaranteed to be even close to the original continuous time signal.
■ Calude Shanon came up with a theorem to do the sampling and
interpolation ‘right’.
■ Then Nyquist also worked a lot in this regard and developed a sampling
theorem.
The sampling theorem
■ The sampling theorem states, that the original continuous time signal can be
reconstructed from its samples exactly,
– when the highest frequency (denoted as ) present in the signal (seen as composition
of sinusoids) is lower than a half of the sampling frequency i.e.

Or
– We must sample at a rate , at least twice that of the highest frequency in the analog
signal Specifically, for having spectral content extending up to , we must choose
whereis equivalent to
– When this condition is met, we can reconstruct the underlying continuous time
signal exactly by a process known as sinc-interpolation
– P.S: half of the sampling rate is also known as “Nyquist frequency ”.
Aliasing
■ When the above condition is not met, but we sample and reconstruct the signal anyway,
an odd phenomenon occurs - it’s called “Aliasing”.

Fig 2.3: Aliasing: the sinusoid at is aliased into another sinusoid at , at a sample rate because
both produce the same sequence of samples
When does alias appear?

■ The general rule is: whenever we sample a sinusoid of a frequency above half the
sample-rate but below the sample rate), then its alias will appear at a frequency of or
■ We’ll see some example to demonstrate the phenomenon of aliasing in class.