Scribd Logo
Upload

Welcome to Scribd!

  • 28 views

    Digital Signal Processing: Solved HW For Day 8

    Uploaded by

    Cuau Suarez
    The document discusses solutions to questions about digital signal processing concepts: 1) Zero phase filtering involves time reversing and filtering a signal twice to produce an output with zero phase delay. 2) The modulation theorem can be used to design a high-pass filter from a low-pass filter and sinusoid generator. 3) The Fourier transform of a weighted time function can be expressed in terms of the Fourier transform of the weighting function using the convolution theorem. 4) The output of an ideal low-pass filter will always be infinite in length if it is non-zero, since the filter has an infinite impulse response.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Digital Signal Processing: Solved HW For Day 8

    Uploaded by

    Cuau Suarez
    28 views9 pages
    The document discusses solutions to questions about digital signal processing concepts: 1) Zero phase filtering involves time reversing and filtering a signal twice to produce an output with zero phase delay. 2) The modulation theorem can be used to design a high-pass filter from a low-pass filter and sinusoid generator. 3) The Fourier transform of a weighted time function can be expressed in terms of the Fourier transform of the weighting function using the convolution theorem. 4) The output of an ideal low-pass filter will always be infinite in length if it is non-zero, since the filter has an infinite impulse response.

    Original Description:

    DSP HW8

    Original Title

    solvedHW8_250514

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document discusses solutions to questions about digital signal processing concepts: 1) Zero phase filtering involves time reversing and filtering a signal twice to produce an output with zero phase delay. 2) The modulation theorem can be used to design a high-pass filter from a low-pass filter and sinusoid generator. 3) The Fourier transform of a weighted time function can be expressed in terms of the Fourier transform of the weighting function using the convolution theorem. 4) The output of an ideal low-pass filter will always be infinite in length if it is non-zero, since the filter has an infinite impulse response.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    28 views9 pages

    Digital Signal Processing: Solved HW For Day 8

    Uploaded by

    Cuau Suarez
    The document discusses solutions to questions about digital signal processing concepts: 1) Zero phase filtering involves time reversing and filtering a signal twice to produce an output with zero phase delay. 2) The modulation theorem can be used to design a high-pass filter from a low-pass filter and sinusoid generator. 3) The Fourier transform of a weighted time function can be expressed in terms of the Fourier transform of the weighting function using the convolution theorem. 4) The output of an ideal low-pass filter will always be infinite in length if it is non-zero, since the filter has an infinite impulse response.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 9

    Digital Signal Processing

    Solved HW for Day 8


    Question 1: Zero Phase filtering

    Let R be the time reversal operator for sequences.

    R{x[n]} = x[−n]

    Let H be a linear time invariant system. What can you say about the following system?

    y [n] = R{H{R{H{x[n]}}}}

    1
    Solution of question 1
    ◮ Let X (e jω ) be the DTFT of the sequence x(n) and H(e jω ) be the frequency response of
    the LTI operator H.
    ◮ By the time reversal property of DTFT, the DTFT of x(−n) is X (e −jω ). Since x(n) is
    real, the DTFT of x(−n) is X ∗ (e jω )

    H{x(n)} = H(e jω )X (e jω )
    R{H{x(n)}} = H ∗ (e jω )X ∗ (e jω )
    H{R{H{x(n)}}} = H(e jω )H ∗ (e jω )X ∗ (e jω )
    R{H{R{H{x(n)}}} = H ∗ (e jω )H(e jω )X (e jω )
    = |H(e jω )|2 X (e jω )
    The above system is a linear, time invariant system with zero phase delay.
    2
    Question 2: Modulation theorem

    Given an ideal low pass filter Hlp (ω) with cut-off frequency ωc and a sinusoid generator at
    frequency π2 , design an ideal high-pass filter Hhp (ω) with cut-off frequency π − ωc . You are
    required to write Hhp with respect to Hlp and a sinusoid.

    3
    Solution of question 2

    ◮ By using the modulation theorem we can see that a high-pass filter with cut-off frequency
    at π − ωc can be written as
    hhp (n) = cos(πn)hlp (n)
    ◮ Since we are only given a sinusoid generator of frequency 2,
    π

    π
    hhp (n) = (2(cos( n))2 − 1)hlp (n)
    2

    4
    Question 3: Convolution Theorem

    Let 
    1 0 ≤ n ≤ M,
    wR [n] =
    0 elsewhere.
    and

     n 0 ≤ n ≤ M/2,
    wT [n] = M − n M/2 < n ≤ M, .
    0 elsewhere.

    Express the Fourier transform WT of wT [n] with respect to the fourier transform WR of wR [n].

    5
    Solution of question 3

    ◮ Let ŵT (n) = wR (n) ⋆ wR (n)



     n 0 ≤ n ≤ M,
    ŵT [n] = 2M − n M < n ≤ 2M,
    0 elsewhere.

    ◮ Thus wT (n) = 12 ŵT (2n)

    ◮ By convolution property, ŴT = WR WR


    ◮ Using scaling property,
    1
    WT = ŴT (ω/2)
    4
    1
    WT = WR (ω/2)WR (ω/2)
    4
    6
    Question 4: Infinite impulse response

    Let H(ω) denote the system response of an ideal low-pass filter. How should the input
    sequence x[n] be, so that the output sequence is finite length and non zero?

    7
    Solution of question 4

    ◮ The impulse response of an ideal low pass filter is,


    ωb ω 
    b
    h(n) = sinc n
    2π 2π
    which is of infinite length.
    ◮ Since any output sequence of the ideal low-pass filter can be written as the sum of scaled
    and shifted versions of this signal, any output sequence is of infinite length, as long as it
    is non-zero.

    You might also like