Scribd
Upload
8 views21 pages

Understanding Signal Aliasing and Sampling

lecture

Uploaded by

Minh Huỳnh
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
8 views21 pages

Understanding Signal Aliasing and Sampling

lecture

Uploaded by

Minh Huỳnh
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

Tutorial 2

• Paper
• Pen
• Calculator
• sampling period T = 1/Fs

• sampled signal x(n) = x(nT) = x(t=nT)

• x(t) with frequency Fx  aliased signal xa(t) with


frequency Fa = Fx + [Link] (k  0)
• x(t) with frequency Fx  aliased signal xa(t) with frequency Fa = Fx + [Link] (k  0)
• F1 = 2, F2 = 6, F3 = 10 (Hz)
• F1aliased = 10 (or -6, 18, …), …
•  xaliased(t) = 10sin(2..10.t) + 10sin(12..t) + 5sin(20..t)
• xaliased(t) x(t) ?
• xaliased(n) = x(n) ?
• x(n) = 10sin(4..n.T) + 10sin(12..n.T) + 5sin(20..n.T)
• xaliased(n) = 10sin(2..10.n.T) + 10sin(12..n.T) + 5sin(20..n.T)
• T = 1/8
• 10sin(4..n.T) = 10sin(4..n/8)= 10sin(.n/2)
• 10sin(2..10.n.T) = 10sin(2..10.n/8)=10sin(2..10.n/8 – 2..n)=10sin(.n/2)
Q1 (answer)
• Fs = 8Hz
• xa1(t) = 10sin(20t) + 10sin(12t) + 5sin(20t)
• xa2(t) = 10sin(36t) + 10sin(12t) + 5sin(20t)

• Fs = 14Hz
• xa1(t) = 10sin(32t) + 10sin(12t) + 5sin(20t)
• xa2(t) = 10sin(60t) + 10sin(12t) + 5sin(20t)
• Nyquist reconstructed frequency range: [-Fs/2  Fs/2]

• Sampling signal x(t) with frequency F  reconstructed signal xa(t) with


frequency Fa = (F + [Link]) [-Fs/2  Fs/2]
Q2 (answer)
• x(t) = sin(6t) + 2sin(6t).cos(4t) = sin(6t) + sin(10t) + sin(2t)

• Fs = 4
• Nyquist reconstructed frequency range: [-2  2]
• fa1 = (3 + k14)  [-2  2] = -1
• fa2 = (5 + k24)  [-2  2] = 1
• fa3 = (1 + k34)  [-2  2] = 1

• Reconstructed signal xa(t) = sin(-2t) + sin(2t) + sin(2t) = sin(2t)


• Sampling period T = 1/Fs

• x(n) = xa(t = nT)


Q3 (answer)
• Sampling period T

• x(n) = xa(t = n.T)


Q4 (answer)
Q4 (answer)
• CT Fourier Transform of xa(t) = cos(w0.t) = cos([Link].f0.t):
• Xa(w) = pi.(w + w0) + pi.(w - w0) {or Xa(f) = 0.5(f + f0) + 0.5(f - f0) }

• Sampling period T
• DT Fourier Transform of xd(n) = xa(t=nT) = cos(wn.n)
• Xd(w) = pi.(w + wn) + pi.(w - wn) but wn [-pi pi]
Q5 (answer a)
Q5 (answer)
Q5 (answer b)
• T = 2ms
• x(n) = cos([Link].n.2/1000) = cos(1,[Link].n) = cos(1,[Link].n – [Link].n) =
cos(-0,[Link].n) = cos(0,[Link].n)
• Xd(w) = pi.(w – 0,[Link]) + pi.(w + 0,[Link])
= pi.(w + 0,[Link]) + pi.(w - 0,[Link])
Q5 (answer c)
• Tmax = ? (no aliasing)

• wn = pi  [Link] = pi  Tmax = pi/w0 = pi/([Link]) = 1/800


• HdB = 20.log10(H)  H = 10^(HdB/20)

• H = 1  HdB = 0 dB

• 1 octave = 2 times, 2 octaves = 4 times, 3 octaves = 8 times, …


• 1 decade = 10 times, 2 decades = 100 times, 3 decades = 1000 times,
…
Q6 (answer)
Q6 (answer)
Q6 (answer)

You might also like