EEG signal analysis, explained.
Short, technical explainers on the parts of EEG analysis that actually matter: frequency bands, band power, the EDF format, and the spectral features people compute from it. Written for researchers, engineers, and students — not marketing copy.
An action potential. One neuron, one spike, ~1 millisecond.
What EEG measures is not this spike. It's the slower voltage change it causes downstream.
Tens of thousands of aligned neurons, summing.
By the time it reaches the scalp: tens of microvolts.
One channel. One continuous voltage over time.
Everything below is about what you can read from that.
EEG Frequency Bands
Delta through gamma — what each band means, and why the boundaries between them are conventions, not constants.
EEG Band Power
Absolute vs relative band power, how it's computed from the PSD, and why the denominator you choose changes everything.
How to Analyze EDF Files
The European Data Format, its structure, and the details that break your analysis if you ignore them (units, sample rates, montages).
Spectral Features of EEG
Spectral edge frequency, median frequency, spectral entropy, line length, and Hjorth parameters — what each captures.