Noise in all its appearances:

Noise is the name we normally give to something that we don't like, we don't want, we don't need, and we wish to get rid of. Noise can come from improper handling of signals or it can occur naturally as a part of nature. In contrast to most "musical" sounds, which have a definite pitch and a predominantly harmonic spectrum, noise consists of a theoretically infinite and continuous range of frequencies. The waveform consists not of regular periodic cycles but of random fluctuations of amplitude. In the present context, the bandwidth of the noise may be considered limited to that of human hearing. As will be seen, it is possible to speak of noise limited to a variety of bandwidths.

Noise description:
White noise: Shot noise:
Pink noise: Popcorn noise or Burst noise:
Red noise also called Brown noise: Generation- Recombination (g-r) noise:
Blue noise: Avalanche noise:
Violet noise or Purple noise: Noise Figure (NF) and Noise temperature
Black noise: Calculating with noise unity:
Semiconductor noise: Related software links:
Johnson noise or Thermal noise: Related links of noise articles:
Flicker noise or 1/f noise: Related Books about noise:

The Federal Standard 1037C Telecommunications: Glossary of Telecommunication Terms defines four noise colours (white, pink, blue & black) and is considered the official source. No official standard could be found for the others. White noise is so named because it is analogous to white light in that it contains all audible frequencies distributed uniformly throughout the spectrum. Passing white light through a prism (a form of filtering) breaks it down into a range of colours. Examination shows that red light is characterized by the longer wavelengths of light, i.e., the lower frequency region. Similarly, "pink noise" has higher energy in the low frequencies, hence the somewhat tongue-in-cheek term.

It is usual to use analogies to visible light when discussing different types of noise. Thus, noise, which has equal intensity at all frequencies, is referred to as "white noise". Because the human ear is more sensitive to high than to low frequencies, white noise will be heard as a high hissing sound. By passing white noise through a filter, different "colours" can be obtained. "Pink noise" results if white noise is passed through a mild low-pass filter. Whereas white noise has equal power at all frequencies, pink noise is defined as having equal power in each octave band (corresponding more closely to the response of the ear). Thus the power varies inversely with frequency - for this reason it is often referred to as  1/f noise. Similarly "red noise" is referred to as 1/f2 (Squared) noise, the high frequencies being much more attenuated than in pink noise.
(Source - Richard Dobson, A Dictionary of Electronic and Computer Music Technology, Oxford University Press, 1992).

White noise:
White noise, like the color white, contains all the elements that other colors of noise are composed of. White noise is truly random in nature and will not correlate. It contains equal energy at all frequencies, (0 dB/oct reference noise with equal power density) . From an energy standpoint; white noise has constant power per hertz (also referred to as unit bandwidth), i.e., at every frequency there is the same amount of power. This noise is useful for dithering to eliminate quantization distortions, music synthesis, and general audio effects, such as chorusing. White noise has the same distribution of power for all frequencies, so there is the same amount of power between 0 and 500 Hz, 500 and 1,000 Hz or 20,000 and 20,500 Hz Therefore the plot of noise power vs. frequency is not flat, but shows a 3 dB rise in amplitude per octave of frequency change. Due to this rising frequency characteristic, white noise sounds very bright and lacking in low frequencies.

Sound of White noise


Pink noise:
Pink noise is evident in all forms of engineering and science from solid-state circuits to astrophysics and music. Unlike white noise, it bears a logarithmic characteristic and as such represents a psycho acoustic equivalent of white noise sweetened for human ears. This is the signal used to test speakers and set equalization in theaters and other public places. When you tune up your home multimedia system, the noise used to drive the speakers for the volume settings is probably pink noise. Pink noise has the same distribution of power for each octave, so the power between 0.5 Hz and 1 Hz is the same as between 5,000 Hz and 10,000 Hz. Since power is proportional to amplitude squared, the energy per Hz will decrease at higher frequencies. (-3 dB/oct decreasing noise density with equal power per octave).

Sound of Pink noise


Red noise also called Brown noise:
Red noise got its name after a connection with red light which is on the low end of the visible light spectrum. Red noise is white noise run through an integrator with frequency response of -6 dB/octave so low frequencies are emphasized and highs are attenuated. The power spectrum of red noise is proportional to the square of the magnitude of the filter ( 1/(f^2) ). (-6 dB/oct decreasing noise density with equal power per octave).
Most amount of low frequency energy or power; used in oceanography.

Sound of Red or Brown noise


Blue noise also called Azure noise:
Blue noise, the inverse of "pink" noise, provides a (+3 dB/oct increasing noise density with equal power per octave). In image processing, the expression blue noise is used for random perturbations favouring high over low frequencies. (sometimes also called 1/f noise , where f is the frequency)

Sound of Blue noise


Violet also called Purple noise:
Violet noise also called Purple noise is the inverse of "Brown noise" provides (+6 dB/oct increasing noise density with equal power per octave).

Sound of Violet or Purple noise


Black noise:
Black noise; noise that has a frequency spectrum of predominately zero power level over all frequencies except for a few narrow bands or spikes. Note: An example of black noise in a facsimile transmission system is the spectrum that might be obtained when scanning a black area in which there are a few random white spots. Thus, in the time domain, a few random pulses occur while scanning.
Black noise
is silence,
(zero power density with a few random spikes allowed).


Semiconductor noise:
This part of the noise page showed how noise behaves itself in electronic parts, like linear components (resistors, capacitors, inductors etc.) and non-linear components (Bipolar-, FET- and MOSFET's semiconductors etc.). The following noises are the most relevant noises. The noise items are derived form mathematical formulas, witch for my point of view to difficult and not interesting enough. In the chapter "Related links of noise articles" at the end of this chapter, you will find interesting web-links to several noise related sites, there you could find the more mathematical background of noise and his behaviour.

Johnson noise Johnson noise or Thermal noise, caused by interactions of the charge carriers (electrons/holes) with the semiconductor lattice.
Flicker noise Flicker noise (1/f noise) is typically due to recombination effects at the semiconductor surface (where there are lots of traps), especially at the gate-oxide interface in MOS transistors. The high "DC" component of 1/f noise is due to charge that is more or less permanently trapped in the oxide of an MOS transistor.
Shot noise Shot noise, which occurs when carriers cross junctions or other energy barriers independently and at random.
Popcorn noise Popcorn noise (Burst noise), primarily found in integrated circuit audio amplifiers that exhibit a sizzling, frying hot-grease kind of sound, similar to popcorn popping. Found to be due to manufacturing defects in the form of metallic impurities in the junctions, often caused by dirty fabrication lines.
(GR) noise Generation-recombination (g-r) noise, in which electrons or holes are captured or released by "traps" in the semiconductor lattice. This type of noise is also predominant in MOS transistors, except that carriers can be captured or released at sites in the insulating gate oxide.
Avalanche noise Avalanche noise, produced by Zener or avalanche break-down in a pn junction.
Noise Figure Noise Figure (NF) and Noise temperature both quality factors of amplifier or semiconductors.
Noise Unity Calculating with noise unity, this sub-chapter describes the varied ways of noise unities and how to handle these.


1) Johnson noise or Thermal noise:
A conductor in thermal equilibrium with its surroundings shows, at its terminals, an open-circuit voltage or short-circuit current fluctuation. It is a form of white noise. For example, a simple resistor hooked up to nothing generates noise, and the larger the resistor value the greater the noise. It is called thermal noise or Johnson noise and results from the motion of electron charge of the atoms making up the resistor (called thermal agitation, which is caused by heat - the hotter the resistor, the noisier. [After John Bertrand Johnson (1887-1970), Swedish-born American physicist who first observed thermal noise while at Bell Labs in 1927, publishing his findings as "Thermal agitation of electricity in conductors," Phys.
Rev., vol. 32, pp. 97-109, 1928.]

This is Noise cause by heating the electrons of a resistor part of a component. The behaviour of the noise is white. The bandwidth is the crucial factor in this.

Voltage noise

Vnoise= mean square root open circuit noise current voltage from resistor R
K = Constant of Boltzmann, (1.38 x10-23 J/0K)
T = Absolute Temperature of the Resistor (0K) in degrees Celsius (0K=0C+273.16)
B= Bandwidth (Hz) or Δf
R= Resistance (

Johnson noise should not be confused with the additional noise voltage created by the effect of resistance fluctuations when an externally applied current flows through a resistor. This "excess noise" has a 1/f spectrum and is heavily dependent on the actual construction of the resistor.

Return Return to semiconductor noise

2) Flicker noise also called 1/f noise or Excess:
Though no universal mechanism has been identified for flicker noise or 1/f noise, it is the most ubiquitous form of noise in nature.
Flicker Noise shows up in Resistors, when it is called “Excess noise”, since this noise is in addition to what is expected from thermal noise considerations. It is found that a resistor exhibits 1/f noise only when there is DC current flowing through it, with the noise increasing with the current. This is widely observed in carbon composition resistors, and the source of noise has been attributed to the formation and extinction of micro-arcs among neighbouring carbon granules.

Noise whose amplitude varies inversely with frequency. Mainly used in solid-state physics to describe noise with 1/f behaviour, such as the noise resulting from impurities in the conducting channel, generation and recombination noise due to base current in transistors, etc. Noise with a power spectrum of 1/f, intermediate to White noise and Brown noise, and sometimes also called 1/f noise.

Flicker Noise is one name given to noise which has a frequency distribution proportional to 1/fα with α close to unity. First seen in tubes where it gets its name due to the apparent flickering of the filament glow. There are various known causes, some possible theories for others and some totally unknown. Other names include Excess Noise, Low-Frequency Noise and 1/f noise for the case of α = 1.

Return Return to semiconductor noise

3) Shot noise:
Another noise mechanism, known as shot Noise was first described and explained by W. Schottky in 1918. The term “shot” is not a corruption of “Schottky”; it is simply that if one hook up an audio system to a source of shot noise biased at low currents, the resulting sound is much that like of buck-shot (pellets) dropping into a hard surface. The fundamental basis for shot noise is the granular nature of electronic charge. Two conditions must be satisfied for shot noise to occur. (1) There must be a direct current flow and (2) there must also be a potential barrier over which the charge carriers hop. The second condition implies that ordinary linear resisters do not generate shot noise, despite the quantized nature of the electronic charge.
The fact that charge comes in discrete bundles means that there are discontinuous pulses of current every time an electron hops an energy barrier. It is the randomness of the arrival times that give rise to the whiteness of shot noise. Shot noise is given by the formula:

Shot noise

Inoise= mean-square root circuit current through R
q=1.6*10-19C (electron charge)

Idc=DC current through R
R= resistance W
B= bandwidth Hz or Δf
f=frequency interval

Shot noise also is ideally white, and has amplitude that possesses a Gaussian distribution. The requirement for a potential barrier implies that shot noise will be only associated with non-linear devices, although not all non-linear devices exhibit shot noise. For example, whereas both the base and collector currents are sources of shot noise in a bipolar transistor because potential barriers are definitely involved there, only the DC gate leakage current of FETs contributes shot noise.

Return Return to semiconductor noise

4) Popcorn noise also called Burst noise:
Burst Noise (Popcorn Noise) is a type of low-frequency noise which varies as 1/f2 at higher frequencies. The mechanism is not fully understood, but the source is related to the presence of heavy-metal ion contamination, such as gold. These ions create deep-level energy traps (near the mid-bandgap). Burst Noise is so named for the fact that an oscilloscope trace of this type of noise shows burst of noise on a number (2 or more) of discrete levels. The repetition rate or the noise pulse is in the audio frequency range and produces a “popping” sound when played through a speaker, leading to its alternate name. Recently seen in very small area MOSFETs with very clean gate oxides (1 or 2 trap energies). Random Telegraph Noise is another name for this case.

It was first observed in point-contact diodes, but has also been seen in ordinary junction and tunnel diodes, some types of resistors and both discrete and integrated circuit junction transistors. This kind of noise is characterized by bi-modal, and hence non-Gaussian amplitude distribution. That is, the noise switches between two or more discrete values at random times.

Return Return to semiconductor noise

5) Generation-recombination (g-r) noise:
g-r Noise (generator-recombination) occurs whenever free carriers are generated and recombine in a semiconductor material. Usually small at room temperature, can be seen at liquid Nitrogen temperature (77K). This process has a characteristic time constant with a spectrum that is constant at low frequencies and varies as 1/f2 at higher frequencies.

The generation-recombination (g-r) process is a natural part of all semiconductor behaviour. In the semiconductor, carriers are freed from association with a particular atom by a generation process, which is necessary for either of the conduction mechanisms to occur (i.e. drift and diffusion). This leaves \uncovered" donor or acceptors which will occasionally trap a passing carrier. Because of the thermal energy of the crystal lattice, the trapped carrier will be freed again after only a short time. This process is a series of independent discrete events. Each event causes fluctuation in the number of free carriers leading to a fluctuation in the material resistance. If a d.c. current is passed through the material, a fluctuating voltage related to the fluctuating resistance will appear across the device. This generation- recombination noise will not appear if there is no d.c. current, is a function of current, however it is not produced by the current and is a low-frequency noise.

Return Return to semiconductor noise

6) Avalanche Noise:
Produced by Zener or avalanche break-down in a pn junction. The holes and electrons in the depletion region must acquire enough energy to create hole-election pairs by colliding with the lattice. This process is multiplicative and results in the production of a random series of noise spikes. This noise is what you get in a MOSFET with significant substrate current. It has a white spectrum.

Return Return to semiconductor noise

7) Noise Figure (NF) and Noise temperature:
The Noise Figure (NF) of an amplifier is the ratio in Decibels of the output of the real amplifier to the output of a ""noiseless" amplifier of the same gain, with a resistor Rs connected across the amplifier's input terminals in each case. In formula:

(Vn)2 is the mean square noise voltage per Hertz contributed by the amplifier, with a noiseless resistor of value Rs connected across its input. Noise Figure is a numeric value in dB to compare other amplifiers of semiconductors with each other. The NF is most of the time available from the manufactory of the semiconductor component as a set of graphs (NF versus f) or (NF versus Rs). A NF specification is always a optimum of Rs versus Ic (current through a transistor this could also be current through a FET Id). When SN+R will be calculated you have to convert the NF factor as follows:

Vn = Voltage(rms) over Rs
K = Constant of Boltzmann, (1.38 x10-23 J/0K)
T = Absolute Temperature of the Resistor (0K) in degrees Celsius (0K=0C+273.16)
Rs= Resistance (

Noise Temperature:
Instead of Noise Figure (NF) sometimes Noise Temperature is used as a quality factor of noise behaviour. Both methods are giving the same information namely the Flicker-, 1/f- or Excess Noise contribution of the amplifier when driven by a signal source of impedance Rs. When Noise Figure (NF) is used,  we have to deal with the following circuit:

Now the circuit with Noise Temperature as noise figure:

In the first figure (NF), Rs is ideal by a T is 0K (-273.160C) saying otherwise noiseless free Rs.
In the second figure, Rs is is now temperature dependent but the amplifier is ideal (noiseless free), Tn is now the Noise Temperature of the Amplifier for a source impedance Rs. The next formulas can be used to convert (NF to Tn) or (Tn to NF):

Tn = Noise Temperature
T = Ambient temperature 390K (273.16+16.840C)
Generally, the better the noise figure, the lower the temperature of the amplifier. In the book The Art of Electronics (ISBN:
ISBN 0-521-37095-7) the noise discussion will go further and explains how to design noiseless arm Amplifiers.

Return Return to semiconductor noise

8) Calculating with noise unity:
In technical literature, noise will be discussed in several unities, like V/(Hz)2 , V2/Hz, V, or V2 .When we speak about noise in V(rms) the following formula apply:

Vn = Noise Voltage(rms) in a bandwidth B
K = Constant of Boltzmann, (1.38 x10-23 J/0K)
T = Absolute Temperature of the Resistor (0K) in degrees Celsius (0K=0C+273.16)
R= resistance
B= bandwidth Hz or Δf

f=frequency interval

The unity could easily convert, see table below:

Formula: Unity:

vn = rms noise voltage in a bandwidth B. (Johnson noise)
Vn = rms noise voltage in a bandwidth B. (Shot noise)

Return Return to semiconductor noise

Related software links:
Tinnitus Masker Pro generates White, Pink, Brown , Purple and Blue noise. You can mix and match combinations and record the output straight to a wav or mp3 file, ready to create your own custom white noise masking CDs or cassettes to listen to anywhere. There is a 14 day trail version downloadable on the following site. 

Audacity is a free audio editor. You can record sounds, play sounds, import and export WAV, AIFF, Ogg Vorbis, and MP3 files, and more. Audacity is being developed by a team of volunteers under the open-source model. There is also the possibility to join the Audacity community and improve the software with your ideas.


Related links of the noise articles: Types.htm


Related Books about noise:

  Noise in Solid State Devices and Circuits, by Aldert. van der Ziel, John Wiley & Sons Inc. 1986. ISBN 0-471-83234-0
  The Art of Electronics, by Paul Horowitz and Winfield Hill, Cambridge University Press, 1989. ISBN 0-521-37095-7 (hardback)