Quantum Internet

Bell Basis Measurement

Also called Bell-state measurement, Bell-state analysis

A Bell basis measurement projects two qubits onto one of the four maximally entangled Bell states, enabling quantum teleportation, entanglement swapping, and superdense coding.

Bloch sphere

A Bell basis measurement asks which of the four Bell states a two-qubit pair occupies. The answer is always one of the four, and the measurement leaves the pair in that state. Because those states are maximally entangled, this is a joint measurement: it cannot be replaced by two separate single-qubit measurements without losing the information it is after.

The four Bell states form a complete orthonormal basis for two qubits. Any two-qubit state can be written as a superposition of them:

∣Φ+⟩=∣00⟩+∣11⟩2|\Phi^+\rangle = \frac{|00\rangle + |11\rangle}{\sqrt{2}}

∣Φ−⟩=∣00⟩−∣11⟩2|\Phi^-\rangle = \frac{|00\rangle - |11\rangle}{\sqrt{2}}

∣Ψ+⟩=∣01⟩+∣10⟩2|\Psi^+\rangle = \frac{|01\rangle + |10\rangle}{\sqrt{2}}

∣Ψ−⟩=∣01⟩−∣10⟩2|\Psi^-\rangle = \frac{|01\rangle - |10\rangle}{\sqrt{2}}

Circuit implementation

The usual circuit is the inverse of Bell-pair preparation: a CNOT (first qubit as control, second as target), then a Hadamard on the first qubit, then a computational-basis measurement on both.

q0:  ─────●──[H]──[M]──  bit 0
          │
q1:  ────[X]──────[M]──  bit 1

CNOT plus Hadamard rotates the Bell basis onto the computational basis. Each Bell state lands on a distinct two-bit string:

  • ∣Φ+⟩\lvert\Phi^+\rangle measures as 0000
  • ∣Ψ+⟩\lvert\Psi^+\rangle measures as 0101
  • ∣Φ−⟩\lvert\Phi^-\rangle measures as 1010
  • ∣Ψ−⟩\lvert\Psi^-\rangle measures as 1111

Bit 0 (the qubit that took the Hadamard) is the phase bit: plus versus minus. Bit 1 is the parity bit: Φ\Phi (even) versus Ψ\Psi (odd). All four states are distinguishable with certainty on any platform that can run that circuit.

Role in teleportation and entanglement swapping

Quantum teleportation uses a Bell measurement as its central step.

  1. Alice holds an unknown qubit ∣ψ⟩|\psi\rangle and one half of a shared Bell pair.
  2. She performs a Bell measurement on those two qubits.
  3. The two classical bits of the outcome go to Bob.
  4. Bob applies the matching Pauli correction to his half of the pair and recovers ∣ψ⟩|\psi\rangle.

No quantum channel carries ∣ψ⟩|\psi\rangle itself. The state moves through the pre-shared entanglement plus the classical bits.

Entanglement swapping is the same protocol on two separate Bell pairs. A middle node holds one qubit from each pair and measures them in the Bell basis. That projects the two remaining qubits (held by parties that never met) into a Bell state. Quantum repeaters use this to stretch entanglement across a chain of links.

Hardware challenges

On matter qubits (trapped ions, superconducting circuits, NV centres) the circuit above is deterministic: those platforms have strong two-qubit gates, so all four Bell states can be told apart.

Photons are harder. A photonic CNOT needs a strong photon-photon nonlinearity that linear optics does not provide. With only beam splitters and single-photon detectors, a Bell measurement can distinguish ∣Ψ+⟩|\Psi^+\rangle and ∣Ψ−⟩|\Psi^-\rangle with certainty, and the other two outcomes are ambiguous. That is the linear-optics Bell-state analysis bound: at most 50% success if the goal is to identify all four states.

Hyperentanglement (polarization plus time-bin, for example) and auxiliary entangled photons can raise the success probability. Complete, deterministic discrimination in linear optics is still an open engineering problem, which is why photonic networks lean so hard on postselection and why matter-photon interfaces matter for repeaters.

See also