1

DIGITAL SIGNAL PROCESSING

Lecture 1 - Chapter 1

Classification of Signals: Continuous-Time verses Discrete-Time Signals

Continuous time or analog signals are signals that are defined for every value of a < t < b, where (a, b) can be ( , + ), i.e., x (t) = e-|t| or x (t ) = cos(t ). .

Discrete-time signals are defined at discrete-time instants and between the two discrete time instants are undefined but are not zero. They can be obtained either by sampling analog signals or they can be discrete in nature like discrete measurement signals.

A discrete-time signal having a set of discrete values is called a digital signal. Note that sampling an analog signal produces a discrete-time signal. Then quantization of its values produces a digital signal.

Deterministic versus Random Signals

Any signal that can be uniquely described by an explicit mathematical expression or a well-defined rule is called deterministic. The past, present and future of a deterministic signal are known with certainty. Otherwise, it is called Random and its properties is explained by statistical techniques.

Review of Sinusoids in Continuous and Discrete Time x a (t ) = A cos(t + )

xa (t + Tp ) = x a (t ) , T p =

= A cos(2Ft + ) = 2F

<t <

1 : fundamental period. Increasing F means increasing F

oscillation in time domain. F = 0 corresponds to Tp = .

2 Also, for complex exponential signals, x a (t ) = Ae j ( t+ ) . A sinusoidal signal then can also be expressed as

xa (t ) = A cos( t + ) =

1 1 Ae j( t + ) + Ae j ( t + ) 2 2
x (n ) = A cos( 0 n + ) = 2 f

Discrete-Time Sinusoid Signals

x (n ) = A cos(2 f 0 n + ) < n < A few important differences between continuous sinusoid and discrete sinusoids: 1) A discrete-time sinusoid is periodic only if its frequency is a rational number. By default, x (n + N ) = x(n ) for all n if x(n) is periodic. The smallest N is called Fundamental Period. x (n + N ) = A cos(2 f 0 (n + N ) + ) = cos(2 f 0 n + ) This relationship is true if and only if 2f 0 N = 2 K f0 = K : a rational number. N

To determine the period N of a periodic discrete time sinusoid, we express f as two relatively prime numbers. Observe that a small change in frequency can result in a 1 large change in period. For example, x1 (n ) = A cos 2 n + = A cos( n + ) its 2 period is N1 = K 30 =2= . f1 60

31 31 Now consider x2 (n) = A cos 2 n + , f 2 = = 0. 517 N 2 = 60 60 60 2. An analog F ( ,+ ) maps to

1 1 f or equivalently to or in 2 2 1 or = + . To see 2

other words, the highest rate of oscillation occurs at f = +

what happens for 2 consider 1 = 0 and 2 = 2 - 0 . When 1 varies between to 2, then 2 varies between and 0. Now x1 (n ) = A cos1 n = A cos 0 n

3 x2 (n) = A cos( 2 )n = A cos(2 0 )n =* A cos( o n ) = A cos on = x1 (n ) * this is only true because n is an integer value, i.e., x(n) is a discrete signal. Hence, x2 (n ) = x1 (n) is an alias because < F < + maps only to by sampling.

1 1 f 2 2

Analog to Digital Conversion (A/D)

sampling (sampling rate) quantization x (n ) = xa (nTs ), where < n < = xa (t ) t =nTs

Sampling:

t = nTs =

n , Fs = Sampling rate (Frequency) (Hz) Fs

Consider an analog sinusoid: xa (t ) = A cos(2 Ft + )

F x (n ) = xa (nTs ) = A cos(2 F nTs + ) = A cos 2 n + Fs

Now recall that < F < maps to

1 1 f 2 2
Nyquist rate

1 F 1 f = 2 Fs 2

FFs 2Fmax 2Fmax s

Example 1:

xa1 (t ) = cos 20 t

xa 2 (t ) = cos 100 t F2 = 50 Hz

F1 = 10 Hz

If we digitize both of these signals with Fs = 40Hz, then

20 n x1 (n ) = xa1 (nTs ) = xa1 = cos n = cos n 40 2 40 100 5 n x2 (n ) = xa 2 = cos n = cos n = cos 2 + n 40 2 2 40 = cos n = x1 (n )!! 2
Therefore, by this sampling rate x2(n) has become same as x1(n), which is an aliasing error. Equivalently, in this case, the frequency of 50 Hz is an alias of 10 Hz by sampling

4 rate of 40 Hz. Furthermore, all frequencies (F1 + 40K) are aliases of F1. Hence, do not use the Nyquist rate blindly. Example 2: xa (t ) = 3 cos 50 t + 10 sin 300 t cos 100 t

14 4 2 3
x1

14243 1 24 4 4 4 3
x2 x3

Nyquist rate? f1 = 25, f2 = 150, f3 = 50 Hz Fmax = 150 Hz Fs = 300 Hz?! Problem: with Fx = 300 Hz, x2 (n ) = xa 2 = 10 sin n = 0 all the time! F s If it had a phase shift o < < n , then it would have been fine, but it is best to choose a higher sampling rate.

Sampling

xa (t )
xa(t)
x

P(t)

x(n)

Ts

x(n)

x p (t ) = x a (t ) p (t )

p(t ) = (t nTs )

x (n ) = x p (t ) | t = nT = x (nTs ) (t nTs )
x

1 [ X a ( )* P ( )] 2 2 + P( ) = ( k s ) Ts Ks X p ( ) = X p ( ) = 1 Ts

X ( k )
a s

Therefore, Xp() is a periodic function of shifted the Xa().

5 X ( ) Obviously, if s 2 M there is no
Xp()

aliasing and the signal can be reconstructed accurately.

In practice however, generating very narrow impulse is very difficult. Therefore, the practical way for sampling is zero-order hold. Such a system samples xa(t) at a given sampling instant and holds that value until the succeeding sampling instant. Reconstruction of xr(t) from the output of this system requires a cascade of low-pass filters or a non-constant gain of LPF.

P(t)

H0(t)

It is as if x a (t )
|Hr()|

xo (t )

xp(t)

-s /2

s /2 <Hr() xr(t)

hr (t )

-s /2

H r ( )

2
s /2