The basics of RF noise
Noise is random by nature, extending across the frequency spectrum in a variety of forms. According to the frequency distribution, there are three categories of noise:
- White noise – white noise affects all frequencies in an equal manner, spreading from zero frequency in an upward fashion with a flat amplitude. The name is fitting as it reflects the fact that white light essentially contains all colors in the spectrum and thus frequencies equally, just as white noise does with sound.
- Pink noise – Unlike white noise, pink noise doesn’t have a flat response. Instead, it contains more frequency bands than others. Due to the nature of pink noise, the power density decreases as frequency increases. Pink noise derives its name from the fact that red light sits at the lower end of the light spectrum and its power density favors lower frequencies.
- Band-limited noise – As noise passes through a circuit or filter it can have its frequency limited.
The effects of electronic / RF noise
Noise can have a variety of effects on systems. For example, amplitude noise can cause data errors to occur or even mask signals. Obviously, systems free of noise will have the best performance, though in most instances there are acceptable levels of noise that systems can handle while still being able to operate at optimal levels.
Types of RF/electronic noise
RF noise can be generated by a variety of mechanisms in a number of ways. Furthermore, the way RF noise is categorized is completely dependent on the way it’s generated:
- Phase noise – Due to the nature of phase noise it is visible to radio frequencies amongst other types of signals. Phase noise appears in the form of perturbations on the signal or phase jitter which manifest in the form of sidebands that spread out from both sides of the carrier or signal.
- Thermal noise – Thermal noise, also commonly referred to as Johnson Nyquist or simply Johnson noise occurs as a result of the thermal agitation of charge carriers in a conductor. The higher the temperature, the more agitated the charge carriers become which in turn causes the level of noise to rise.
Thermal noise is significant in low noise amplifiers. To reduce thermal noise, high-performance amplifiers have to be operated at extremely low temperatures.
- Flicker noise, 1/f noise – Flicker noise occurs in nearly every type of electronic device. This occurs for a variety of reasons; each of which is related to the direct current flow. Flicker noise has a frequency spectrum that steadily falls off into higher frequencies.
- Avalanche noise – Avalanche noise occurs when a junction diode operates near the point of avalanche breakdown. This is a phenomenon that takes place in semiconductor junctions when the carriers located within a high voltage gradient builds up an adequate amount of energy to unseat any additional carriers by way of physical impact. Avalanche noise is the direct result of this impact.
- Shot noise – Shot noise occurs due to fluctuations in electrical current (which is often time dependent). Short noise can be noticed particularly in semiconductor devices (such as tunnel junctions, p-n diodes, and Schottky barrier diodes).
- Burst noise – Burst noise occurs in some circuits where a semiconductor, when operated, experiences a sudden impulse. As a result, audio circuits pick up “burst noise” which is also referred to as “popcorn noise.”
Ultimately, electrical or RF noise plays a key role in any system. In fact, there are many instances in which it can determine how a system will perform. For instance, the noise that a radio receiver picks up can limit the sensitivity of the radio itself.
In cameras, noise can actually be seen, especially in settings where the lighting is low. Overall, the noise performance of all electronic equipment can be extremely important to its overall performance levels.
Noise is also an extremely important factor for most types of electronic circuits as well. In most cases noise levels are important, and the actual noise levels must be measured to ensure that they fall within acceptable limits. Otherwise, measurements will need to be made to determine whether or not the levels can be improved in any capacity.
In light of this information, it is very necessary to create standardized methods of measuring, assessing and specifying noise levels. In this manner, the RF noise level can be measured which makes comparisons possible to similar items of test equipment and similar circuits.
The noise levels in electronic circuits can be specified in a variety of ways. However, the manner in which it is specified is entirely contingent upon the application of the circuit in question.
Radio receivers play a big role in how noise is specified. When it comes to radio receivers, noise specifications relate to the sensitivity of the receiver and includes specifications such as noise figure and noise ratio.
The primary factor that may limit receiver sensitivity is noise. Furthermore, receiver sensitivity specifications tend to revolve around noise specifications. The primary receiver specifications are noise figure, SINAD, and signal to noise ratio. Each specification looks to the receiver performance in relation to noise.
Techniques for noise measurement
There are a wide variety of ways to measure noise in an electronic or RF system. Several specialized meters are used to measure SINAD, noise figure, and other like figures. Standard test equipment may also be factored into the equation when appropriate.
- Spectrum analyzer – Modern day spectrum analyzers will more likely than not have the ability to measure noise levels. To ensure measurements are done correctly the settings of the analyzer should be configured properly to ensure optimum conditions are set. The settings used will be largely bedependent on the type of analyzer used to perform the measurement.
- Meter method – Meters can be used (along with other elements) to properly measure noise levels. A simple noise measurement system will display an ordinary measurement circuit. Any and all noise that’s generated by the unit while it’s under stress is amplified until it reaches a suitable level.
The meter used to perform the measurement should have an averaging capability – this is because noise levels are random by nature and will often vary. Be mindful that most digital meters have an averaging capability whereas analog meters will average any changes made.
Precautions when measuring noise
When it comes to measuring noise levels, spectrum analyzers are the easiest way to accomplish thisgoal. Spectrum analysis is capable of determining the level of noise in any given bandwidth. There are several factors to keep in mind during this process:
- Spectrum analyzer noise performance – The noise performance of the spectrum must be “better” than the actual noise being measured. If not, then the readings reflected from the test will only display the performance of the spectrum analyzer and not that of the unit being tested.
- Filter shape – Because the spectrum analyzer requires a finite band to transition from its pass-band to its stop-band, the shape of the spectrum analyzer must be taken into account when the calculation for the noise in a given bandwidth is being made.