Superdense coding

Superdense coding uses a shared Bell state to communicate two classical bits by transmitting one qubit. The sender selects one of four orthogonal Bell states with a local Pauli operator, and the receiver performs a Bell-basis measurement. For an ideal shared Bell pair, the transmitted qubit alone has a maximally mixed state for every message and reveals no information about the chosen encoding.

With phase/parity labels $(u,v)$ encoded by $X^v{Z}^u$, a resource $a|{\Phi }^{+}⟩+b|{\Phi }^{-}⟩$ gives intended-label probability $|a{|}^2$ and phase-label-flip probability $|b{|}^2$ under a Bell-basis measurement. This follows because the local Pauli operator maps the two orthogonal components into the two Bell states whose phase label differs. The resource is a coherent pure state, even though its measured label statistics form a classical noisy channel.

If a wrong Bell-label partner occurs with probability $ϵ$ and differs from the intended two-bit label in $w$ positions, its contribution to the error probability of a uniformly selected bit is $ϵw/2$. Relabeling the Bell states can change $w$ while preserving the physical wrong-state probability. Thus symbol error probability and bit error probability are different quantities.

For four uniformly distributed Bell states, measure one qubit along Bloch vector $\mathbf{n}$ and the other along $z$. The sign within each ${\Phi }^{\pm }$ or ${\Psi }^{\pm }$ pair is invisible and the displayed mutual information is at most one bit. A local Hadamard gate before a computational detector changes $n_z$ from one to zero and destroys the remaining information. An entangling Bell decoder falls outside this product-measurement restriction and recovers two bits.