Overview
Harry Nyquist (1889–1976) established at Bell Labs the sampling condition that bears his name: a signal containing no frequencies above some limit is completely determined by samples taken at more than twice that limit. Shannon later formalised it. It is the reason every digital audio and video sample rate is the number it is.
What You Need
- Lived: 1889–1976, Sweden, worked in the United States
- Anchor year: 1928 — the sampling criterion
- Strand: overlooked — a Bell Labs name in every spec sheet
Steps
The problem as it stood
Telegraph engineers needed to know how fast pulses could be sent through a channel of given bandwidth before they smeared into each other. The underlying question — how much information a bandlimited channel can carry — had no rigorous answer.
What he actually did
Nyquist established the relationship between bandwidth and the maximum rate of distinguishable pulses, and the corresponding sampling condition: to represent a signal bandlimited to B, you need more than 2B samples per second.
How it worked
Sample too slowly and frequencies above half the sample rate do not disappear — they fold back into the range below it and appear as false lower frequencies. That is aliasing, and it is irreversible: once folded, the false content is indistinguishable from real content. Hence the anti-alias filter before every analogue-to-digital converter.
What it made possible
All digital audio and video. 44.1 kHz exists because human hearing extends to roughly 20 kHz and a little headroom is needed for the filter; 48 kHz is the professional equivalent. Video sampling, codec design and every converter obey the same constraint.
What happened to him
He spent his career at AT&T and Bell Labs, contributing also to thermal noise theory — Johnson-Nyquist noise — and feedback stability. He retired in 1954 and died in 1976.
Where the credit landed
Nyquist's name survives in the "Nyquist frequency" and the "Nyquist rate", so his attribution is unusually intact — though the theorem is often called Nyquist-Shannon, and the work is Bell Labs work whose patents belonged to the company. Almost nobody who cites the Nyquist rate knows anything else about him.
Pro Tips
- Aliasing is irreversible — filter before sampling, not after.
- 44.1 kHz follows from ~20 kHz hearing plus filter headroom.
- The wagon-wheel effect in film is aliasing in the time domain.
Knowledge Base
What You'll Learn
Every digital format in this corpus obeys one inequality.
Why aliasing is worse than losing information
If undersampling simply discarded high frequencies the result would be dull but honest. Instead those frequencies fold down and masquerade as lower ones that were never in the original — a 30 kHz tone sampled at 44.1 kHz appears as 14.1 kHz, audible and wrong. Nothing downstream can remove it, because it is now indistinguishable from real signal. This is why the anti-alias filter is placed before the converter and is not optional.
Where This Fits
This guide covers one specific part of the history of media technology. The wider picture — how each link in the chain from capture through transmission to display was actually built, who built it, and why the credit so often landed somewhere else — is in A History of Broadcast Technology: The Chain From Capture to Screen, which frames the discipline as a whole and links out to the detailed guides underneath it, including this one. If you are starting from scratch rather than solving a specific problem, read that first and come back here.
FAQ
Q: Why is CD audio 44.1 kHz?
A: Human hearing extends to roughly 20 kHz, so the Nyquist condition requires more than 40 kHz. The extra margin allows a practical anti-alias filter; the precise figure also suited the video recorders used to store early digital masters.
Q: What is aliasing?
A: When a signal is sampled too slowly, frequencies above half the sample rate fold back and appear as false lower frequencies, indistinguishable from genuine content. It cannot be undone after sampling.