5.

NOISE
Noise is an electronic signal that is a mixture of many random frequencies at
many amplitudes that get added to a radio or information signal as it is transmitted
from one place to another or as it is processed. Noise is not the same as
interference from other information signals. If the noise level is high enough
and/or the signal is weak enough, the noise can completely obliterate the original
signal. Noise that occurs in transmitting digital data causes bit errors and can
result in information being garbled or lost. The noise level in a system is
proportional to temperature and bandwidth, and to the amount of current flowing
in a component, the gain of the circuit, and the resistance of the circuit. Increasing
any of these factors increases noise. Therefore, low noise is best obtained by using
low-gain circuits, low direct current, low resistance values, and narrow
bandwidths. Keeping the temperature low can also help.

5.1 Signal to Noise Ratio

The signal-to-noise (S/N) ratio, also designated SNR, indicates the relative
strengths of the signal and the noise in a communication system. The stronger the
signal and the weaker the noise, the higher the S/N ratio. If the signal is weak and
the noise is strong, the S/N ratio will be low and reception will be unreliable.
Signals can be expressed in terms of voltage or power. The S/N ratio is computed
by using either voltage or power values:
5.2 External Noise
External noise comes from sources over which we have little or no control
industrial, atmospheric, or space. The amplitude varies over a wide range, as does
the frequency. One can say that noise in general contains all frequencies, varying
randomly. This is generally known as white noise.
5.2.1 Industrial Noise. Industrial noise is produced by manufactured
equipment, such as automotive ignition systems, electric motors, and generators.
Any electrical equipment that causes high voltages or currents to be switched
produces transients that create noise. Noise pulses of large amplitude occur
whenever a motor or other inductive device is turned on or off. The resulting
transients are extremely large in amplitude and rich in random harmonics.
Fluorescent and other forms of gas-filled lights are another common source of
industrial noise.
5.2.2 Atmospheric Noise
The electrical disturbances that occur naturally in the Earth’s atmosphere are
another source of noise. Atmospheric noise is often referred to as static. Static
usually comes from lightning, the electric discharges that occur between clouds
or between the Earth and clouds. Huge static charges build up on the clouds, and
when the potential difference is great enough, an arc is created and electricity
literally flows through the air. Lightning is very much like the static charges that
we experience during a dry spell in the winter. The voltages involved are,
however, enormous, and these transient electric signals of megawatt power
generate harmonic energy that can travel over extremely long distances.
Like industrial noise, atmospheric noise shows up primarily as amplitude
variations that add to a signal and interfere with it. Atmospheric noise has its
greatest impact on signals at frequencies below 30 MHz.
 5.2.3Extraterrestrial Noise
Extraterrestrial noise, both solar and cosmic, comes from sources in space. One
of the primary sources of extraterrestrial noise is the Sun, which radiates a wide
range of signals in a broad noise spectrum. The noise intensity produced by the
Sun varies with time. In fact, the Sun has a repeatable 11-year noise cycle. During
the peak of the cycle, the Sun produces an immense amount of noise that causes
tremendous radio signal interference, rendering many frequencies unusable for
communication. During other years, the noise is at a lower level.
Noise generated by stars outside our solar system is generally known as cosmic
noise. Although its level is not as great as that produced by the Sun, it is
nevertheless an important source of noise that must be considered due to the great
distances between those stars and Earth. It primarily shows up in the 10-MHz to
1.5-GHz range, causing the greatest disruptions in the 15- to 150-MHz range.
5.3 Internal Noise
Electronic components in a receiver such as resistors, diodes, and transistors are
major sources of internal noise. Internal noise, although it is low level, is often
great enough to interfere with weak signals. The main sources of internal noise in
a receiver are thermal noise, semiconductor noise, and intermodulation distortion.
5.3.1 Thermal Noise
Most internal noise is caused by a phenomenon known as thermal agitation, the
random motion of free electrons in a conductor caused by heat. Increasing the
temperature causes this atomic motion to increase. Since the components are
conductors, the movement of electrons constitutes a current flow that causes a
small voltage to be produced across that component. Electrons traversing a
conductor as current flows experience fleeting impediments in their path as they
encounter the thermally agitated atoms. The apparent resistance of the conductor
thus fluctuates, causing the thermally produced random voltage we call noise.
Thermal agitation is often referred to as white noise or Johnson noise.White noise
contains all frequencies randomly occurring at random amplitudes. A white noise
signal therefore occupies, theoretically at least, infinite bandwidth. Filtered or
band-limited noise is referred to as pink noise.
Noise is a broadband signal containing a wide range of random frequencies. Its
level can be reduced by limiting the bandwidth. If a noise signal is fed into a
selective tuned circuit, many of the noise frequencies are rejected, leading to an
overall decrease in the noise level.
The noise power is proportional to the bandwidth of any circuit to which it is
applied. Filtering can reduce the noise level but does not eliminate it entirely. The
amount of open-circuit noise voltage appearing across a resistor or the input
impedance to a receiver can be calculated using Johnson’s formula.

Thermal noise can also be computed as a power level. Johnson’s formula is then

5.3.2 Semiconductor Noise

Electronic components such as diodes and transistors are major contributors to
noise. In addition to thermal noise, semiconductors produce shot noise, transit-
time noise, and flicker noise. The most common type of semiconductor noise is
shot noise. Current flow in any device is not direct and linear. The current carriers,
electrons or holes, sometimes take random paths from source to destination,
whether the destination is an output element, tube plate, or collector or drain in a
transistor. It is this random movement that produces the shot effect. Shot noise is
also produced by the random movement of electrons or holes across a PN
junction. Shot noise is also white noise in that it contains all frequencies and
amplitudes over
a very wide range. The amplitude of the noise voltage is unpredictable, but it does
follow a Gaussian distribution curve that is a plot of the probability that specific
amplitudes will occur. The amount of shot noise is directly proportional to the
amount of dc bias flowing in a device. The rms noise current in a device 𝐼 is
calculated with the formula

5.3.3 Intermodulation Distortion

Intermodulation distortion results from the generation of new signals and
harmonics caused by circuit nonlinearities. Nonlinearities produce modulation or
heterodyne effects. Any frequencies in the circuit mix together, forming sum and
difference frequencies. When many frequencies are involved, or with pulses or
rectangular waves, the large number of harmonics produces an even larger
number of sum and difference frequencies. When two signals are near the same
frequency, some new sum and difference frequencies are generated by a
nonlinearity, and they can appear inside the bandwidth of the amplifier. In most
cases, such signals cannot be filtered out. As a result, they become interfering
signals to the primary signals to be amplified. They are a form of noise.

Expressing Noise Levels

The noise quality of a receiver can be expressed as in terms of noise figure, noise
factor, noise temperature, and SINAD.
 Noise Factor and Noise Figure
The noise factor is the ratio of the S/N power at the input to the S/N power at the
output. The device under consideration can be the entire
receiver or a single amplifier stage. The noise factor or noise ratio (NR) is
computed with the expression

When the noise factor is expressed in decibels, it is called the noise figure (NF)

NF= 10 log NR dB

Amplifiers and receivers always have more noise at the output than at the input
because of the internal noise, which is added to the signal. The S/N ratio at the
output will be less than the S/N ratio of the input, and so the noise figure will
always be greater than 1. A transistor amplifier in a communication receiver
usually has a noise figure of several decibels. The lower the noise figure, the
better the amplifier or receiver. Noise figures of less than about 2 dB are excellent.

 Noise Temperature
Most of the noise produced in a device is thermal noise, which is directly
proportional to temperature. Therefore, another way to express the noise in an
amplifier or receiver is in terms of noise temperature TN
Noise temperature is expressed in kelvins.
Where

The relationship between noise temperature and NR is given by

  SINAD
Another way of expressing the quality and sensitivity of communication receivers
is SINAD the composite signal plus the noise and distortion divided by noise and
distortion contributed by the receiver. In symbolic form,

Distortion refers to the harmonics present in a signal caused by nonlinearities.

The SINAD ratio is also used to express the sensitivity of a receiver. To obtain
the SINAD ratio, an RF signal modulated by an audio signal is applied to the
input of an amplify er or a receiver. The composite output is then measured,
giving the S +N+D.
The SINAD is a power ratio, and it is almost always expressed in decibels

SINAD is the most often used measure of sensitivity for FM receivers used in
two-way radios.
Noise in Cascaded Stages
Noise has its greatest effect at the input to a receiver simply because that is the
point at which the signal level is lowest. The noise performance of a receiver is
invariably determined in the very first stage of the receiver, usually an RF
amplifier or mixer. Design of these circuits must ensure the use of very low-noise
components, taking into consideration current, resistance, bandwidth, and gain
figures in the circuit. Beyond the first and second stages, noise is basically no
longer a problem.
The formula used to calculate the overall noise performance of a receiver or of
multiple stages of RF amplification, called Friis’ formula, is

Note that the noise ratio is used, rather than the noise figure, and so the gains are
given in power ratios rather than in decibels.