AD/DA conversion

What is the main purpose of deliberately adding a small noise (dither) before AD conversion?

  1. ATo make quantization error resemble noise uncorrelated with the signal, reducing distortion and allowing averaging to improve resolution
  2. BTo lower the sampling frequency
  3. CTo increase the input signal amplitude
Show answer and explanation

Answer: A. To make quantization error resemble noise uncorrelated with the signal, reducing distortion and allowing averaging to improve resolution

When the input is small or changes slowly, the quantization error is correlated with the signal and appears as distortion (harmonics). Adding suitable dither makes the error behave like noise uncorrelated with the signal, reducing distortion. With averaging, values finer than 1 LSB can also be estimated.

  • A. Correct. It decorrelates the quantization error and reduces distortion.
  • B. Dither does not change the sampling frequency.
  • C. Dither is a very small noise, not a means of increasing amplitude.
Practice this topic Back to topic

More AD/DA conversion questions

  1. What is the effect of raising the order of the modulator in a ΔΣ AD converter (for example, from first to second order)?
  2. In a time-interleaved ADC that alternates between several ADCs, what problem is caused by mismatches in offset, gain and timing between the individual ADCs?
  3. Which correctly describes the "offset binary" code used in ADCs that handle both positive and negative voltages?
  4. How many output levels can an 8-bit AD converter represent?
  5. What sampling frequency is needed to sample an analog signal containing components up to 20 kHz without aliasing?