The Quantum Fourier Transform is the discrete Fourier transform implemented as a unitary operation on qubit amplitudes. Acting on a computational basis state it produces:
QFT|k⟩ = (1/√N) Σ_{j=0}^{N-1} e^{ 2πi jk/N } |j⟩
IQFT|k⟩ = (1/√N) Σ_{j=0}^{N-1} e^{-2πi jk/N } |j⟩
|c_j| = 1/√N for every j (equal magnitude)
arg(c_j) = 2π jk/N (mod 2π) (linearly winding phase)
Each needle in the ring is one basis state |j⟩. Stem height = |c_j| (probability amplitude magnitude); the horizontal arrow atop it points at the complex phase arg(c_j), colour-coded the same way (hue = phase angle). Before the transform, all probability mass sits in a single spike at |k⟩ with phase 0. After Apply Transform, the amplitude spreads to every output state with identical magnitude 1/√N, but the phase advances by a fixed step Δφ = 2πk/N between adjacent states — exactly the "clock hand" winding that makes the QFT a frequency detector.
- Qubits n — register size; N = 2ⁿ basis states (rendered on the ring).
- Basis index k — which computational basis state |k⟩ is fed into the transform.
- Forward / Inverse — sign of the exponent; the inverse QFT winds phase the opposite direction and is the step used to read out a phase-estimation register.
- ΣP — sum of |c_j|² over all states, shown live as a sanity check that the transform stays unitary (always 1).
Real-world relevance: the QFT is the engine inside Shor's factoring algorithm and quantum phase estimation — after a controlled-unitary encodes a periodic phase into the register, the QFT converts that period into a measurable peak in the output distribution, giving quantum computers their speed-up for period-finding problems.