Sampling Theorem

Why it matters

The Sampling Theorem is fundamental in digital signal processing, allowing engineers to convert continuous-time signals into discrete-time signals without losing information. This is essential for digital communication systems, audio processing, and any application where analog signals need to be digitized.

Key ideas

• Sampling: The process of converting a continuous-time signal into a discrete-time signal by taking samples at regular intervals.
• Nyquist Rate: The minimum sampling rate required to avoid aliasing, which is twice the highest frequency present in the signal.
• Aliasing: A phenomenon that occurs when a signal is undersampled, causing different signals to become indistinguishable.
• Reconstruction: The process of converting a discrete-time signal back into a continuous-time signal, ideally recovering the original signal.

Formulas

• Nyquist rate = 2 * f_m
  • f_s: Sampling frequency (Hz)
  • f_m: Maximum frequency of the signal (Hz)

Assume a baseband signal bandlimited to |f| ≤ f_m and ideal uniform sampling/reconstruction; use f_s > 2f_m for a general guarantee. The Nyquist frequency is f_s/2, a different quantity. Practical signals require an anti-alias filter and transition-band margin.

Worked example

Given: A continuous-time signal with a maximum frequency of 1 kHz.

1. Determine the Nyquist rate.

   • Formula: f_s = 2 * f_m
   • Calculation: f_s = 2 * 1 kHz = 2 kHz

2. Choose a sampling frequency greater than the Nyquist rate.

   • Example: f_s = 2.5 kHz

3. Verify that the chosen sampling frequency avoids aliasing.

   • Since f_s = 2.5 kHz > 2 kHz, aliasing is avoided.

Final Answer: Nyquist rate = 2 kHz; the chosen 2.5 kHz exceeds it for the assumed strictly bandlimited signal. Exact equality can lose a sinusoid at the boundary depending on phase.

Common mistakes

• Confusing the Nyquist rate with the Nyquist frequency.
• Forgetting to sample at a rate higher than the Nyquist rate.
• Misidentifying the maximum frequency of the signal.

For GATE EE

Questions often involve calculating the Nyquist rate, identifying aliasing in sampled signals, and determining appropriate sampling frequencies. Practice problems that require understanding the relationship between sampling frequency and signal reconstruction.

Quick check

1. What is the Nyquist rate for a signal with a maximum frequency of 500 Hz?
2. What happens if a signal is sampled below its Nyquist rate?
3. How can aliasing be avoided in signal sampling?

Answers: 1. 1000 Hz 2. Aliasing occurs 3. Sample at a rate higher than the Nyquist rate.