Superdense Coding with GHZ and Quantum Key Distribution with W in the ZX-calculus

Quantum entanglement is a key resource in many quantum protocols, such as quantum teleportation and quantum cryptography. Yet entanglement makes protocols presented in Dirac notation difficult to verify. This is why Coecke and Duncan have introduced a diagrammatic language for quantum protocols, called the ZX-calculus. This diagrammatic notation is both intuitive and formally rigorous. It is a simple, graphical, high level language that emphasises the composition of systems and naturally captures the essentials of quantum mechanics. In the author’s MSc thesis it has been shown for over 25 quantum protocols that the ZX-calculus provides a relatively easy and more intuitive presentation. Moreover, the author embarked on the task to apply categorical quantum mechanics on quantum security; earlier works did not touch anything but Bennett and Brassard’s quantum key distribution protocol, BB84. Superdense coding with the Greenberger-Horne-Zeilinger state and quantum key distribution with the W-state are presented in the ZX-calculus in this paper.

💡 Research Summary

The paper presents a diagrammatic treatment of two quantum communication protocols—super‑dense coding using the Greenberger‑Horne‑Zeilinger (GHZ) state and quantum key distribution (QKD) using the W‑state—within the ZX‑calculus, a graphical language rooted in categorical quantum mechanics. After motivating the need for a more intuitive formalism, the authors briefly review the ZX‑calculus: Z‑spiders (green nodes) and X‑spiders (red nodes) together with a small set of rewrite rules (fusion, bialgebra, Hopf, etc.) are sufficient to represent any unitary, measurement, and classical control operation.

In the first part, the GHZ state (|\text{GHZ}\rangle = (|000\rangle+|111\rangle)/\sqrt{2}) is expressed as a simple three‑legged Z‑spider. Alice’s encoding of two classical bits is modelled by attaching appropriate X‑ or Z‑spider phases to her two qubits; the four possible encodings (I, X, Z, XZ) correspond to the four combinations of phase 0 or (\pi) on the green and red spiders. The ZX‑diagram makes clear that these operations commute with the GHZ entangling spider, so the encoded state remains a valid GHZ‑type diagram. Bob’s decoding consists of applying the inverse spider phases, fusing the diagram back to the original GHZ spider, and finally measuring in the computational basis. The authors show, using only the spider‑fusion and colour‑change rules, that the diagram reduces deterministically to the two‑bit classical outcome, thereby proving correctness without any matrix algebra.

The second part introduces the W‑state (|W\rangle = (|001\rangle+|010\rangle+|100\rangle)/\sqrt{3}). Unlike GHZ, W has an asymmetric entanglement structure: tracing out any qubit leaves the remaining pair still entangled. The authors construct a ZX‑diagram for (|W\rangle) by decomposing it into a combination of Z‑ and X‑spiders together with ancillary phases. In the QKD protocol, Alice, Bob, and Charlie each hold one qubit of the shared W‑state. They perform local Z‑basis measurements, represented by attaching a measurement spider to each leg, and then exchange the classical outcomes over a public channel. The key bits are derived from parity checks that exploit the fact that exactly one of the three measurement results is “1”.

Security analysis is carried out by inserting an eavesdropper’s intervention as an extra spider (or a pair of spiders) into the diagram. Using the ZX rewrite rules, the authors demonstrate that any non‑trivial intervention introduces an unwanted phase that propagates to the measurement layer, causing a detectable error rate in the parity check. This graphical proof mirrors the standard information‑theoretic argument but is achieved solely through diagrammatic rewriting, highlighting the ZX‑calculus’s capacity to capture security properties.

A comparative discussion follows. GHZ‑based super‑dense coding achieves the maximal classical‑bit‑per‑qubit transmission rate (two bits per qubit) but is fragile: loss of any qubit destroys the entire encoded information. W‑based QKD, by contrast, tolerates the loss of a single qubit without breaking the remaining entanglement, offering robustness at the cost of a lower key‑generation rate. The ZX‑calculus makes these trade‑offs visually apparent: the GHZ diagram is a single high‑arity spider, while the W diagram consists of a more distributed network of spiders, each reflecting different resilience properties.

In conclusion, the paper demonstrates that the ZX‑calculus provides a high‑level, compositional framework for both designing and verifying quantum communication protocols. By translating algebraic steps into intuitive graphical rewrites, it simplifies correctness proofs, clarifies the role of entanglement structure, and offers a transparent method for security analysis. The authors argue that such diagrammatic reasoning will become an essential tool for future research in quantum cryptography and broader quantum information science.