1 · Why this mathematics matters
Characterise electrical noise, correlation, stationarity and power spectral density.
2 · Prerequisites and learning approach
Prerequisites: Random variables and Fourier analysis. Begin with the physical meaning, identify every symbol and unit, study the fixed-scale graph, then solve and verify the examples.
3 · Symbols and units
4 · Concepts and textbook formulas
1. Mean and stationarity
A wide-sense stationary process has constant mean and autocorrelation that depends only on time separation.

How to read it: identify the output quantity first, then the variables, parameters, units and assumptions. Test zero, unity and limiting values before using the equation in a design.
2. Autocorrelation
Autocorrelation measures similarity with a delayed copy and reveals periodicity and memory.

How to read it: identify the output quantity first, then the variables, parameters, units and assumptions. Test zero, unity and limiting values before using the equation in a design.
3. Power spectral density
The Wiener-Khinchin relation connects autocorrelation and power spectrum.

How to read it: identify the output quantity first, then the variables, parameters, units and assumptions. Test zero, unity and limiting values before using the equation in a design.
4. Noise models
White noise has flat PSD; thermal-noise voltage grows with resistance, bandwidth and temperature.

How to read it: identify the output quantity first, then the variables, parameters, units and assumptions. Test zero, unity and limiting values before using the equation in a design.
5 · Step-by-step worked examples
Example 1: Foundation
Doubling measurement bandwidth doubles noise power and increases RMS noise by sqrt(2).
Verification: check the unit, sign, order of magnitude and physical trend.
Example 2: Thermal noise
Given: R=10 kohm, T=300 K and B=100 kHz.

Engineering meaning: Noise voltage grows with the square root of resistance, temperature and bandwidth.
Example 3: Averaging uncorrelated noise
Given: Average N independent measurements.

Engineering meaning: Averaging sixteen independent samples reduces RMS noise by a factor of four.
6 · EE/ECE applications
- receiver noise analysis — use the governing equation to predict parameter sensitivity, confirm a simulation and interpret measured behaviour.
- clock jitter — use the governing equation to predict parameter sensitivity, confirm a simulation and interpret measured behaviour.
- sensor filtering — use the governing equation to predict parameter sensitivity, confirm a simulation and interpret measured behaviour.
- communication channels — use the governing equation to predict parameter sensitivity, confirm a simulation and interpret measured behaviour.
7 · Common mistakes
- Assuming one noise waveform represents the complete process.
- Confusing amplitude spectrum with power spectral density.
8 · Practice problems
- Interpret the autocorrelation width of a noise process.
- Relate autocorrelation and PSD using Wiener–Khinchin.