Red noise is a term for randomness in signals that is not totally uncorrelated (e.g., white noise) but has a particular frequency distribution with more of the lower frequencies (so-named because the color red comprises the lower frequencies of visible light). The term is often used specifically for the case where the power at a particular frequency is related to the reciprocal of the square of the frequency (1/f²). It is also called Brown noise or Brownian noise because Brownian motion matches it.
The term is used in discussion of signal detection. One example is the detection of extra-solar planets through astrometry, for which measurement noise is due to equipment and atmospheric anomalies.
For random noise with some other frequency distributions, the analogous terms: pink noise, blue noise, and violet noise are used:
|term||tendency||relation to frequency|
|white noise||same regardless of frequency||constant|
|pink noise||falls with frequency||~1/f|
|red noise||more pronounced fall||~1/f²|
|blue noise||rises with frequency||~f|
|violet noise||more pronounced rise||~f²|