Noise (signal processing)

From Wikipedia, the free encyclopedia

In signal processing, noise is a general term for unwanted (and, in general, unknown) modifications that a signal may suffer during capture, storage, transmission, processing, or conversion.[1]

Sometimes the word is also used to mean signals that are random (unpredictable) and carry no useful information; even if they are not interfering with other signals or may have been introduced intentionally, as in comfort noise.

Noise reduction, the recovery of the original signal from the noise-corrupted one, is a very common goal in the design of signal processing systems, especially filters. The mathematical limits for noise removal are set by information theory.

Types of noise[]

Signal processing noise can be classified by its statistical properties (sometimes called the "color" of the noise) and by how it modifies the intended signal:

  • Additive noise, gets added to the intended signal
  • Multiplicative noise, multiplies or modulates the intended signal
  • Quantization error, due to conversion from continuous to discrete values
  • Poisson noise, typical of signals that are rates of discrete events
  • Shot noise, e.g. caused by static electricity discharge
  • Transient noise, a short pulse followed by decaying oscillations
  • Burst noise, powerful but only during short intervals
  • Phase noise, random time shifts in a signal

Noise in specific kinds of signals[]

Noise may arise in signals of interest to various scientific and technical fields, often with specific features:

  • Noise (audio), such as "hiss" or "hum", in audio signals
    • Background noise, due to spurious sounds during signal capture
    • Comfort noise, added to voice communications to fill silent gaps
    • Electromagnetically induced noise, audible noise due to electromagnetic vibrations in systems involving electromagnetic fields
  • Noise (video), such as "snow"
  • Noise (radio), such as "static", in radio transmissions
  • Image noise, affects images, usually digital ones
    • Salt and pepper noise or spike noise, scattered very dark or very light pixels
    • Fixed pattern noise, that is tied to pixel sensors
    • Shadow noise, made visible by increasing brightness or contrast
    • Speckle noise, typical of radar imaging and interferograms
    • Film grain in analog photography
    • Compression artifacts or "mosquito noise" around edges in JPEG and other formats
  • Noise (electronics) in electrical signals
  • Synaptic noise, observed in neuroscience
  • Neuronal noise, observed in neuroscience
  • Transcriptional noise in the transcription of genes to proteins
  • Cosmic noise, in radioastronomy
  • Phonon noise in materials science
  • Internet background noise, packets sent to unassigned or inactive IP addresses
  • Fano noise, in particle detectors
  • Mode partition noise in optical cables
  • Seismic noise, spurious ground vibrations in seismology
  • Cosmic microwave background, microwave noise left over from the Big Bang

Measures of noise in signals[]

A long list of noise measures have been defined to measure noise in signal processing: in absolute terms, relative to some standard noise level, or relative to the desired signal level. They include:

Technology for noise in signals[]

Almost every technique and device for signal processing has some connection to noise. Some random examples are:

See also[]

References[]

  1. ^ Vyacheslav Tuzlukov (2010), Signal Processing Noise, Electrical Engineering and Applied Signal Processing Series, CRC Press. 688 pages. ISBN 9781420041118
Retrieved from ""