arXiv:2609.38441
Classical Verifier Position Verification from Non-Local Games
Wen Yu Kon, Fatih Kaleoglu, Kaushik Chakraborty
Abstract
Secure position verification certifies that a remote prover occupies a claimed location, a task that is provably impossible with classical resources alone against colluding adversaries. Most existing quantum position verification schemes require transmitting quantum states between verifiers and provers, and the resulting photon loss makes long-distance verification impractical. Classical verifier position verification (CVPV) schemes with classical verifier-prover communication exist, but rely on complex quantum processes that generate certifiable randomness, placing them outside the reach of near-term hardware. Here we introduce a general compiler that maps any complete-support non-local game with the required quantum advantage into a multi-prover CVPV protocol. This addresses both limitations, with entirely classical verifier-prover communication and all quantum resources confined to shared entanglement and local measurements on the prover devices. Instantiating with the CHSH game enables near-term implementation on existing experimental platforms. We prove finite-size security in the quantum random oracle model for both simultaneous verification of multiple provers and verification of a single prover with multiple prover devices. Notably, security in the multi-prover setting is governed not by the certified randomness of the joint provers' outputs, but by certified blind local randomness that depends critically on the spatial arrangement of provers and verifiers.
AI Summary
The paper asks whether quantum position verification can use entirely classical verifier–prover communication without requiring complex quantum computation at the prover. It constructs a compiler from complete-support non-local games to protocols with multiple separated prover devices sharing entanglement and performing local measurements. Verifiers split challenges into shares, combine them through random-oracle hashes, and check both the game score and consistency of responses delivered within location-dependent deadlines. Finite-size soundness is proved in the quantum random oracle model against sequential adversaries with bounded oracle queries, allowing quantum memory across rounds. Security requires geometric conditions isolating each scored response from foreign inputs and forcing another response copy to be guessed without its generating system. The proof reduces spacetime attacks to non-local-game strategies, derives score–mismatch tradeoffs from multipartite blind local randomness, and applies concentration bounds. When adversaries may occupy all but one prover location, the score must exceed the optimum with one classical player; under stronger protected-region and geometric restrictions, exceeding the fully classical optimum suffices. For CHSH, both models admit the analytical bound $p_{\mathrm{mis}}\geq\frac{8}{9}(\omega-3/4)^2$, with tighter numerical SDP bounds. A GHZ-based attack shows that quantum advantage and global output randomness alone do not ensure security: the accessible side information depends on geometry. Complete-support modification or geometric changes can restore security in the discussed examples. No implementation is reported. The proofs idealize processing delays; quantitative security geometry, positional uncertainty, minimum device separation, and practical timing and transmission imperfections remain open.
Central claim may not hold: The compiler samples independent uniform game inputs rather than the arbitrary joint distribution assumed in its theorems, and this change can eliminate a game's quantum advantage. The QROM soundness penalty incorrectly depends on hash output length instead of the length of an unrevealed challenge share. The near-term secure-implementation claim lacks a completed security analysis incorporating nonzero prover processing time.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.