Topic: Fourier Transform and Its Applications in Signal Analysis
Fourier Transform Fundamentals:
● Purpose: Decompose signals into constituent frequencies.
● Continuous-Time Fourier Transform (CTFT): X(jω)=∫−∞∞x(t)e−jωtdtX(j\omega) =
\int_{-\infty}^{\infty} x(t) e^{-j \omega t} dtX(jω)=∫−∞∞x(t)e−jωtdt
● Inverse CTFT: x(t)=12π∫−∞∞X(jω)ejωtdωx(t) = \frac{1}{2\pi} \int_{-\infty}^{\infty}
X(j\omega) e^{j \omega t} d\omegax(t)=2π1∫−∞∞X(jω)ejωtdω
Properties of Fourier Transform:
1. Linearity:
○ ax1(t)+bx2(t)↔aX1(jω)+bX2(jω)a x_1(t) + b x_2(t) \leftrightarrow a
X_1(j\omega) + b X_2(j\omega)ax1(t)+bx2(t)↔aX1(jω)+bX2(jω)
2. Time Shifting:
○ x(t−t0)↔e−jωt0X(jω)x(t - t_0) \leftrightarrow e^{-j \omega t_0}
X(j\omega)x(t−t0)↔e−jωt0X(jω)
3. Frequency Shifting:
○ ejω0tx(t)↔X(j(ω−ω0))e^{j \omega_0 t} x(t) \leftrightarrow X(j(\omega -
\omega_0))ejω0tx(t)↔X(j(ω−ω0))
4. Scaling:
○ x(at)↔1∣a∣X(jωa)x(a t) \leftrightarrow \frac{1}{|a|} X\left( \frac{j\omega}{a}
\right)x(at)↔∣a∣1X(ajω)
5. Convolution:
○ x(t)∗h(t)↔X(jω)⋅H(jω)x(t) * h(t) \leftrightarrow X(j\omega) \cdot
H(j\omega)x(t)∗h(t)↔X(jω)⋅H(jω)
6. Differentiation in Time Domain:
○ dx(t)dt↔jωX(jω)\frac{d x(t)}{dt} \leftrightarrow j \omega
X(j\omega)dtdx(t)↔jωX(jω)
Applications:
● Filtering:
○ Design filters (low-pass, high-pass, band-pass) by manipulating
X(jω)X(j\omega)X(jω).
● Modulation:
○ Used in communication systems to shift signals in frequency.
● Signal Reconstruction:
○ Sampling and reconstructing signals using the Fourier Transform and the
Sampling Theorem.
Discrete-Time Fourier Transform (DTFT):
● For discrete signals x[n]x[n]x[n]: X(ejω)=∑n=−∞∞x[n]e−jωnX(e^{j\omega}) =
\sum_{n=-\infty}^{\infty} x[n] e^{-j \omega n}X(ejω)=∑n=−∞∞x[n]e−jωn
Discrete Fourier Transform (DFT):
● Definition: X[k]=∑n=0N−1x[n]e−j2πkn/N,k=0,1,...,N−1X[k] = \sum_{n=0}^{N-1} x[n]
e^{-j 2\pi k n / N}, \quad k = 0, 1, ..., N-1X[k]=∑n=0N−1x[n]e−j2πkn/N,k=0,1,...,N−1
● Inverse DFT: x[n]=1N∑k=0N−1X[k]ej2πkn/Nx[n] = \frac{1}{N} \sum_{k=0}^{N-1} X[k]
e^{j 2\pi k n / N}x[n]=N1∑k=0N−1X[k]ej2πkn/N
● Properties:
○ Periodicity: X[k+N]=X[k]X[k + N] = X[k]X[k+N]=X[k]
○ Symmetry: For real-valued signals, X[N−k]=X∗[k]X[N - k] =
X^*[k]X[N−k]=X∗[k]
Fast Fourier Transform (FFT):
● Purpose: Efficient computation of DFT.
● Radix-2 FFT Algorithm:
○ Requires NNN to be a power of 2.
○ Divides the DFT computation into smaller DFTs recursively.
Windowing Techniques:
● Purpose: Minimize spectral leakage in finite-length signals.
● Common Windows:
○ Rectangular, Hamming, Hanning, Blackman.
● Effect on Spectrum:
○ Trade-off between main-lobe width and side-lobe levels.
Spectral Analysis:
● Power Spectral Density (PSD):
○ Describes how the power of a signal is distributed over frequency.
● Estimation Methods:
○ Periodogram: Squared magnitude of the Fourier Transform.
○ Welch's Method: Averages periodograms over segments to reduce
variance.
Sampling and Aliasing:
● Nyquist Criterion:
○ To avoid aliasing, sampling frequency fsf_sfsmust satisfy fs≥2fmaxf_s
\geq 2 f_{max}fs≥2fmax.
● Aliasing:
○ High-frequency components appear as lower frequencies if fsf_sfsis
insufficient.
Example Calculation:
● Signal:
○ x(t)=cos(2πf0t)x(t) = \cos(2\pi f_0 t)x(t)=cos(2πf0t), where f0f_0f0is the
signal frequency.
● Fourier Transform:
○ X(jω)=π[δ(ω−2πf0)+δ(ω+2πf0)]X(j\omega) = \pi [\delta(\omega - 2\pi f_0) +
\delta(\omega + 2\pi f_0)]X(jω)=π[δ(ω−2πf0)+δ(ω+2πf0)]
● Interpretation:
○ Two delta functions at ω=±2πf0\omega = \pm 2\pi f_0ω=±2πf0, representing
the frequency components.
Assignments:
● Lab:
○ Use MATLAB/Python to compute and plot the FFT of a given signal.
○ Analyze the effect of different window functions.
● Reading:
○ Chapter 7 on Discrete-Time Signal Processing.
● Exercises:
○ Problems involving the computation of Fourier Transforms of complex
signals.
○ Application of properties to simplify Fourier Transform calculations.
●
