You are on page 1of 10

6.

011: Signals, Systems &


Inference

Lec 2
Transforms

1
DT convolution to z-transform
(and system function)
X1
y[n] = h[k]x[n - k]
k= 1

n
x[n] = z0 , all n

⇣ X
1 ⌘
y[n] = h[k]z0 k z0n
k= 1
| {z }
H(z0 )
2
DT convolution to DTFT
(and frequency response)
X1
y[n] = h[k]x[n k]
k= 1

x[n] = (ej⌦0 )n , all n, and ⇡ < ⌦0  ⇡

⇣ X
1 ⌘
j⌦0 k j⌦0 n
y[n] = h[k]e (e )
k= 1
| {z }
3 H(ej⌦0 )
Using frequency response to specify
response to sinusoidal inputs

x[n] = A cos(⌦0 n + ✓)

⇣ ⌘
y[n] = |H(e j⌦0
)| A cos ⌦0 n + ✓ + \H(e j⌦0
)

4
Frequency response
(DTFT of unit sample response)
1
X
j⌦ j⌦k
H(e ) = h[k]e
k= 1

When is the above infinite sum well-defined?

Z ⇡
1
h[n] = H(ej⌦ )ej⌦n d⌦
2⇡ ⇡

5
Spectral content of a signal
(DTFT of the signal)

1
X
j⌦ j⌦k
X(e ) = x[k]e
k= 1

Z ⇡
1
x[n] = X(ej⌦ )ej⌦n d⌦
2⇡ ⇡

6
Classes of DT signals that have DTFTs

Absolutely summable (finite “action”, have


continuous spectra)
– or not, but …
Square summable (finite “energy”, spectra
have discontinuities)
– or not, but …
Bounded (finite amplitude, spectra involve
generalized functions like impulses)
7
Convolution in time to
multiplication in frequency

Putting together frequency response,


spectral content, and superposition, we find

y[n] = h ⇤ x[n]
in the time domain translates to

j⌦ j⌦ j⌦
Y (e ) = H(e )X(e )

in the frequency domain.

8
Exercise

Derive the DTFT of


1
X
r[n] = x[k]x[k - n]
k= 1

9
MIT OpenCourseWare
https://ocw.mit.edu

6.011 Signals, Systems and Inference


Spring 2018

For information about citing these materials or our Terms of Use, visit: https://ocw.mit.edu/terms.

10

You might also like