How to handle Noisy Signals in your instrumentation measurements?

Unwanted or irrelevant signals that obstruct the precise measurement of a desired amount or characteristic are referred to as noise in measurements. 

What are the types of noise that occur during measurement?

Depending on the measurement type being used, noise might take many various formsThey are Electrical Noise, Shot Noise, Ground Loops, Thermal Noise, Environmental Noise, Sensor Noise, Quantization Noise, Instrumentation Noise, Atmospheric Noise

Electrical Noise

• Electrical noise may affect electrical measurements because it results from electrical interference sources. 
• It includes sources including power line noise, radio frequency interference (RFI), and electromagnetic interference (EMI). 
• The measured signal may experience random alterations or distortions due to electrical noise. 

Electrical Interference: Nearby power lines, electrical devices, motors, transformers, and other sources of electromagnetic fields can all result in electrical interference. The measurement system may experience undesired signals or noise as a result.

Radio Frequency Interference (RFI): RFI is brought on by radio frequency signals transmitted by electronic devices such as radio transmitters, wireless devices, communication equipment, and other electronic gadgets. RFI can disrupt delicate measurement equipment and contaminate the measured signals.

Thermal Noise

All electrical circuits contain thermal noise, which also goes by the names Johnson noise and white noise and results from the erratically thermal motion of electrons. It has a constant spectrum of frequencies and varies in direct proportion to temperature. Low-level electrical measurements may be impacted by thermal noise, although it can be reduced by using signal amplification or averaging techniques. 

The Root Mean Square (RMS) voltage due to thermal noise Vn, generated for a given a resistance R (ohms) and a bandwidth f (Hz)

v_n = √ (4K_BTR∆f)

Where 

Shot Noise

Due to electrical current’s discrete character, shot noise is a sort of noise that might develop. It results from stochastic variations in an electrical circuit’s electron or photon flow. Shot noise, which is more perceptible at lower current or light levels, is particularly pertinent in low-intensity electrical or optical studies. 

 The Schottky formula determines the root-mean-square value of the shot noise current, 

i_n = √ (2Iq∆B)

where

• I – Current
• q – Charge of an electron
• ΔB – Bandwidth

Ground Loops

When a measurement system has several grounding routes or potential discrepancies between various components, ground loops might develop. Inadvertent electrical currents may be introduced as a result, causing noise and distorted signals.

Environmental Noise

The term “environmental noise” refers to outside influences on measurements, such as ambient vibrations, temperature changes, humidity changes, or audio disruptions. Sensitive measurements, such as those employing accelerometers, temperature sensors, or sound sensors, may be subject to uncertainty or inaccuracy as a result of these circumstances. 

Sensor Noise

The limitations or flaws in the sensing device or transducer that is utilized for measurement give rise to sensor noise. Thermal fluctuations, electrical noise, or restrictions in the sensor’s physical qualities are possible causes. Temperature, pressure, strain, and position readings can all be impacted by sensor noise. 

Quantization noise

Digital measurements that use analog-to-digital converters (ADCs) to turn discrete analogue signals into discrete digital values are subject to quantization noise. It happens as a result of the ADC’s low resolution and appears as errors or signal distortions in the digitized signal. When the signal’s amplitude approaches the ADC’s resolution limit, quantization noise becomes more noticeable. 

Systematic Noise

Non-random or predictable disruptions that consistently affect measurements are referred to as systematic noise. It can be caused by a number of things, including drift over time, nonlinearity in sensors or equipment, calibration mistakes, electronic offsets, and errors in calibration. Measurement biases or inaccuracies caused by systematic noise call for strategies for calibration or correction. 

Atmospheric noise

• Lightning and other nearby natural electrical activity are the main causes of atmospheric noise. 
• Additional sources of electrical noise include fluorescent lighting, high-tension wires, switchgear, electric motors (particularly those with brushes), and automobile ignition systems.
•  Solar noise fluctuates. Coronal discharges and sunspots are two examples. 

Noise Reduction Techniques

Many times, unwanted noise on a signal in a circuit is found. One often seeks a true output of a circuit’s accomplishments while designing one. A noisy altered output signal can be converted to a more theoretical output signal using a variety of noise reduction techniques. 

Faraday cage

Using a Faraday cage can help reduce total noise in a circuit. The Faraday cage as a container that isolates the entire circuit from external power lines and any other signal that might tamper with the real signal. The vast majority of electrostatic and electromagnetic noise is typically blocked by a Faraday cage.

Unwanted capacitive coupling

It can be produced in a circuit by a current flowing through two resistors or any other type of conductor that are close to one another. If this occurs, an AC signal from one area of the circuit may unintentionally find its way to another. Due to the capacitor-like behavior of the two resistors (conductors), AC signals are sent. Capacitive coupling might be desired for other reasons, in which case it would not be regarded as electronic noise. 

Ground Bus

It’s crucial to prevent ground loops when grounding a circuit. When there is a voltage difference between the two ground potentials, ground loops develop. Any location on a ground bus should not have a voltage because ground is considered to be 0V. It would not be a true ground if this were the case. Using a ground bus to connect all the ground wires to the same potential is a viable solution for this problem.

Shielded Cable

Using shielded cables to protect the wires from unwelcome noise frequencies in a sensitive circuit is often a good idea. Since a shielded wire uses a piece of plastic or rubber to enclose the actual wire, it can be compared to a miniature Faraday cage. A conductive metal that is immediately outside the rubber/plastic covering intercepts any noise signal. The noise signal flows directly to ground before reaching the real wire since the conductive metal is grounded. To prevent a ground loop on the shield, it’s crucial to ground it only at one end. 

Twisted pair wiring

By tightly twisting the wires in a circuit, electromagnetic noise is significantly reduced. The size of the loop through which a magnetic field can flow to create a current between the wires can be reduced by twisting the wires. Even if the wires are tightly twisted, there may still be a little loop between them. However, because the wires are twisted, the magnetic field passing through the smaller loops causes a current to flow in each wire in the opposite direction, cancelling out the smaller loops. 

Nozzle filters

When removing a particular noise frequency, notch filters or band-rejection filters are crucial. For instance, the internal power lines of a building typically operate at 60 Hz. A sensitive circuit may occasionally detect this 60 Hz noise through an unwelcome antenna (which might be as basic as a wire in the circuit). The desired signal will be amplified without enhancing the noise at 60 Hz by passing the output through a notch filter at that frequency.  In a way, the noise will be eliminated at the filter’s output. 

Explain noise in Bioinstrumentation Measurement signals 

• In a biological instrumentation system are tampered with noise. When undesirable signals from external sources, such as power cables and electromagnetic radio and television waves, enter the system, interference noise results.By paying close attention to the wiring arrangement of the circuit to decrease coupling effects, this type of noise can be successfully eliminated. 
• Power lines (50 or 60 Hz), fluorescent lights, AM/FM radio broadcasts, computer clock oscillators, laboratory equipment, and mobile phones all produce interference noise.
• Capacitive and/or inductive coupling injects electromagnetic noisefrom sources into the amplifier circuit or into the patient.
•  At the sensor/amplifier interface, noise is produced even by the action potentials brought on by nerve conduction in the patient. 

How noise due to bioinstrumentation is reduced?

• At the A/D converter’s input, filters are utilised to minimise noise and maximise the signal-to-noise (S/N) ratio.
  • A high-pass filter with the cutoff frequency set above the noise frequencies and below the biological signal frequencies removes low-frequency noise (amplifier d.c. offsets, sensor drift, temperature changes, etc.). 
  • A low-pass filter with the cutoff set above the frequencies of the biological signal being monitored and below the noise frequencies reduces high-frequency noise (nerve conduction, radio broadcasts, computers, cell phones, etc.). 
  • Biological signal being recorded typically falls within the 50- or 60-Hz frequency range, power line noise poses a highly challenging problem in biological monitoring.The noise from power lines is frequently reduced with band-stop filters. These band-stop filters have their cutoff frequencies a few Hertz to either side of the notch frequency, which is adjusted to the power line frequency of 50 or 60 Hz.
• Inherent noise is the second kind of signal corruption. Because inherent noise results from random events that are essential to how a circuit’s components work, it can be decreased by smart circuit design. 
• Although intrinsic noise can be diminished, it can never be completely removed.Low-pass filters is used to cut down on high-frequency elements. This filtering method is unable to remove noise signals that fall within the frequency range of the amplified biosignal.

Explain Noises in Digital Multimeter

Phantom voltage in the multi meter is confusing. When there is no input, it makes the voltage in the readout wander. It has many strange noises, namely including thermal, Johnson, Nyquist, or white noise.

Explain Noises in Oscilloscopes 

• In an oscilloscope in FFT or RF mode, it typically manifests as the noise floor, a flickering horizontal line. It manifests as a thickening of the trace in the temporal domain and, in extreme cases, results in a loss of triggering. 
• A fraction of the noise is eliminated by restricting the bandwidth to 250 MHz, and limiting it to 20 MHz is still more efficient. However, a drawback of this noise mitigation method is that it frequently necessitates using the entire bandwidth to display the desired signal. 
• The second method of noise reduction is signal averaging. As a result of each waveform’s unique noise content, noise is then suppressed. Waveforms are not suppressed, assuming they are periodic. 
• Depending on the type of noise, different techniques are used to reduce external noise to the instrument. Example, air carries the electrical noise produced by surrounding devices. The Faraday cage is one method of stopping it. 

List few techniques to handle noisy signals.

Filtering

Removing undesired noise from a signal with a filter is one method of dealing with noisy signals. To eliminate high-frequency or low-frequency noise, filters can be applied digitally or analogously. The cut-off frequency of the filter must be carefully chosen to prevent obliterating important information from the signal. 

Average signal

Another strategy for lowering signal noise is averaging. When a signal is averaged, many measurements of the same signal are taken. Although it might not be helpful in reducing systematic noise, this strategy can help to reduce random noise.

Shielding

Protecting the measurement apparatus and wires from electromagnetic interference is known as shielding. Shielded cables, adequate equipment grounding, and placing the equipment far from electromagnetic interference sources can all help achieve this.

Signal Conditioning

To reduce noise and improve the signal-to-noise ratio, the signal is amplified and filtered. It is possible to utilise amplifiers, attenuators, and filters for signal conditioning.

Frequent Calibration

For the instrumentation to be correct and the measurements to be trustworthy, calibration is necessary. The detection and correction of any flaws that might be creating noisy signals can be aided by routine calibration.

Signal-to-Noise Ratio (SNR)

It’s necessary to take the measuring system’s signal-to-noise ratio into account. The SNR can be improved by boosting the signal or cutting back on the noise. By comparing the power of the signal to the power of the noise, the SNR can be computed.