Protocol balances accuracy and privacy in quantum sensing networks

Researchers at the Technical University of Denmark and Sorbonne Université have introduced and analysed a protocol for distributed quantum sensing that estimates an average phase with precision increasing alongside the total photon number, exhibiting Heisenberg scaling. The work answers the question of whether a network can estimate a global parameter while protecting locally encoded values, with each node encoding a local phase into a shared entangled Gaussian state.

Complete privacy is unattainable for finite squeezing in multi-party settings, but it emerges in the large-squeezing limit. The team further investigated the impact of displacements and optical losses, revealing trade-offs between estimation accuracy and privacy.

Continuous-Variable Network Enables Distributed Quantum Sensing

A network of quantum sensors can now estimate a global parameter while simultaneously shielding locally encoded information, a feat demonstrated through a new continuous-variable protocol. The network’s architecture is central to this advancement; each node encodes information about a local phase rather than measuring any parameter. This localized encoding, combined with the shared entangled state, allows for global estimation without revealing individual node data.

The team’s analysis of the quantum Fisher information matrix (QFIm) reveals a critical distinction between two-mode and multi-mode settings, establishing that a two-mode squeezed state remains completely private, but local knowledge of a party’s phase constitutes a break in privacy, even with finite squeezing and optical loss. Although complete privacy, where all other combinations of phases remain entirely hidden, is unattainable for finite squeezing in multi-party settings, it emerges in the large-squeezing limit.

The researchers quantified privacy using a measure, P(Q,v), introduced in a prior reference, and found that while individual phases are inaccessible, complete privacy is unattainable for finite squeezing in multi-party settings. “Increasing the squeezing causes all QFIm eigenvalues to grow, but the dominant eigenvalue grows faster, asymptotically driving the matrix towards rank 1,” the paper states, explaining how increased squeezing enhances privacy by limiting accessible information to the average phase.

The team benchmarked their protocol against other continuous-variable resource states, demonstrating its competitive performance. They also noted that if a node knows its own phase, it can then infer the other’s phase from the estimated average. This research demonstrates a pathway towards building quantum sensor networks that can collaboratively estimate parameters while safeguarding the privacy of individual nodes.

Multipartite Gaussian State Protocol Encodes Local Phases

A new protocol allows a quantum sensor network to estimate an average phase with precision while simultaneously shielding individual node values from being revealed, a feat previously considered a challenging trade-off. This approach exhibits Heisenberg scaling in the total photon number, meaning the precision of the average phase estimation increases directly with the number of photons employed. The protocol’s core lies in the careful engineering of correlations within the distributed Gaussian state.

By tailoring these correlations, the network confines accessible information to the average phase, effectively suppressing any leakage regarding other combinations of individual phases. Although complete privacy, where all other combinations of phases remain entirely hidden, is unattainable for finite squeezing in multi-party settings, it emerges in the large-squeezing limit.

Quantum Fisher Information Matrix Quantifies Estimation & Privacy

Their work, published this month, introduces and analyses a protocol where a network of quantum sensors can accurately determine an average phase while preventing access to individual phase values encoded at each node. This is achieved through careful manipulation of entanglement within a shared Gaussian state distributed across the network. The study highlights the impact of displacements and optical losses, revealing trade-offs between estimation accuracy and privacy.

The team’s work provides a quantitative framework for assessing privacy in quantum sensor networks, utilizing the QFIm to identify unobservable directions, vectors in space where no information about individual phases can be inferred. This detailed analysis offers a crucial step towards building secure and high-performance distributed quantum sensing systems, with potential applications in areas like precision metrology and secure communication.

Two-Mode Squeezed States Guarantee Complete Privacy

This finding addresses a critical challenge in distributed quantum sensing: achieving high precision in parameter estimation without compromising the confidentiality of individual sensor data. The protocol detailed in the work establishes a pathway toward secure quantum networks capable of collaborative sensing tasks. Each node within the network encodes a local phase onto a shared entangled Gaussian state, a specific configuration that differs from approaches relying on discrete variables.

The network’s ability to conceal individual phase combinations improves as squeezing intensifies. Optical losses, an unavoidable reality in any physical system, also present a trade-off between estimation accuracy and privacy. The work demonstrates that carefully engineered multipartite continuous-variable states can retain complete privacy properties even under realistic conditions.

Privacy Loss with Local Phase Knowledge in Two-Party Setups

Recent work on quantum sensor networks reveals a seemingly counterintuitive result: achieving high precision in estimating a global parameter doesn’t necessarily require sacrificing the privacy of individual sensor readings. For a two-party setup, the researchers found a privacy guarantee, but this complete privacy is contingent on a crucial condition, local knowledge of a party’s phase constitutes a break in privacy. This highlights a nuanced interplay between local knowledge and global privacy within the network. The study also investigates the impact of practical limitations like displacements and optical losses.

Displacements & Optical Losses Impact Estimation Accuracy

Maintaining this delicate balance between estimation precision and privacy is not without its challenges, particularly when considering real-world limitations inherent in quantum networks. The study reveals that displacements, disruptions to the symmetry of the phase space, actively compromise the network’s ability to conceal individual phases. These disturbances introduce asymmetries that allow for the potential inference of local information, eroding the intended privacy safeguards.

Equally impactful are optical losses, which degrade the quality of the entangled state and introduce noise into the system. We further investigate the impact of displacements and optical losses, revealing trade-offs between estimation accuracy and privacy.

This formulation moves beyond simply achieving private estimation and provides a quantifiable way to assess how well the network conceals information about individual phases. The researchers further investigated how displacements and optical losses impact both estimation accuracy and privacy.