Abstract
Quantum gambling — a secure remote two-party protocol which has no classical counterpart — is demonstrated through an optical approach. A photon is prepared by Alice in a superposition state of two potential paths. One path leads to Bob and is split into two parts. The security is confirmed by quantum interference between Alice's path and one part of Bob's path. It is shown that a practical quantum gambling machine can be feasible by this way.
Introduction
In classical remote gambling, trust is the primary hurdle. If Alice (the casino) and Bob (the player) are in different locations, ensuring fairness usually requires a trusted third party. Alice might misrepresent the state of the "betting box," or Bob might lie about the outcome. This paper addresses these challenges by moving the game into the quantum domain, where security is derived from the fundamental laws of physics rather than a middleman's integrity. Quantum states cannot be measured or manipulated without leaving a detectable trace, allowing for a fair game where cheating is naturally penalized.
Experimental Setup
The researchers implemented the protocol using a standard optical bench setup. The "particle" used in the game is a linear-polarized photon (simulated by a He-Ne laser beam). The game involves two potential paths, $|a\rangle$ and $|b\rangle$, representing the betting boxes. Alice prepares the photon in a superposition of these paths using a Half-Wave Plate (HWP $a$) to adjust the "bias" ($\epsilon$).
Bob receives path $|b\rangle$ and uses another HWP ($b_1$) to perform a unitary operation, splitting the state into a new path $|b'\rangle$ with a parameter $\eta$. This serves as his verification move. If Bob does not find the photon in his path, Alice must send her path $|a\rangle$ to him. Quantum interference is then used to verify Alice's honesty. If Bob detects the photon in a specific "punishment" state ($|\phi_b\rangle$ at detector $D_3$), Alice is caught cheating.
Results and Discussion
The experiment's data consists of relative light intensities measured by three detectors, representing three possible outcomes: Bob wins 1 coin ($D_1$), Bob loses 1 coin ($D_2$), or Alice is caught cheating and pays a penalty $R$ ($D_3$). The fairness of the game relies on this pre-agreed punishment $R$. If $R$ is sufficiently high, Alice's optimal strategy is to play fairly ($\epsilon = 0$).
The paper demonstrates a Nash equilibrium in this optical system. For a given penalty $R=5$, the researchers identified an optimal splitting parameter $\tilde{\eta} \approx 0.27$. At this point, Bob's expected gain is stabilized against Alice's choices, ensuring a fair game regardless of her attempt to bias the state.
Furthermore, the researchers evaluated error tolerance. For the protocol to remain secure against device errors, the error rate must be lower than $\sqrt{2/R^3}$. With an experimental error rate of approximately $1/40$, the machine is practical for penalties up to $R < 14.4$.
Conclusion
This work successfully maps an abstract quantum gambling protocol onto a physical optical system, demonstrating its feasibility. While the current experiment used a laser beam ensemble, a truly secure implementation would require a single-photon source to prevent specific attacks. Future work may address environmental sensitivity in long-distance fiber optics and explore more complex quantum games involving multiple players or entangled states.
Summary updated