D.A. Kronberg of the Steklov Mathematical Institute, Russian Academy of Sciences, proposes a new method for perfectly discriminating between subsets of quantum states, even when those states are non-orthogonal and linearly dependent. The work challenges a core tenet of quantum mechanics, which typically limits the success probability of such discrimination to below one. Kronberg demonstrates that by accepting unpredictability in the outcome, a success probability of one becomes achievable, bypassing established quantum restrictions. This approach also offers a potential advantage in quantum cryptography, promising unit sifting probability even with numerous bases and non-orthogonal states.

Quantum Subsets Discrimination Achieves Unit Success Probability

D.A. Kronberg’s achievement relies on a technique called postselection, where attempts resulting in failure are discarded, a strategy similar to probabilistic cloning or unambiguous state discrimination. The approach fundamentally challenges the notion that uncertainty is inherent in discriminating non-orthogonal quantum states, which traditionally limits success probability below one. Previous work focused on maximizing success for a single division, achieving probabilities higher than 1/3 but still falling short of unity, but Kronberg’s method shifts the focus to a broader, randomized approach.

“Thus, the long-standing prohibition of error-free discrimination for non-orthogonal quantum states is not absolute,” Kronberg writes in the published work. This development has direct implications for quantum cryptography, specifically in key sifting procedures. The method proposes an alternative basis reconciliation procedure capable of providing unit sifting probability, even when employing a large number of bases and non-orthogonal states within each basis.

This contrasts with conventional methods that struggle with imperfect discrimination, potentially leading to key loss and reduced security. The research, conducted at the Steklov International Mathematical Center and supported by the Ministry of Science and Higher Education of the Russian Federation, demonstrates that by relaxing the requirement for a predictable division of subsets, a fully reliable discrimination process can be achieved.

Gram Matrices and Pure State Mapping for Discrimination

The core of the technique involves decomposing an input Gram matrix into combinations of Gram matrices representing different divisions of the quantum states, each with an associated probability. This decomposition allows for a mapping onto several possible sets of pure states, effectively sidestepping the usual restrictions on error-free discrimination. The method’s efficacy is particularly evident with equidistant states, where the inner product between any two states equals a constant value ‘c’, defined as (N-2)/(2N-2) for N states.

Achieving conclusive subsets discrimination for these states becomes possible under this framework, provided the inner product meets the specified condition. The number of balanced divisions, those with an equal number of elements in each subset, is determined by the binomial coefficient $C_{N - 1}^{N/2 - 1}$ , a value that dictates the complexity of the decomposition process.

A key element is the use of diagonal matrices to account for potentially unequal probabilities assigned to each division of input states. This allows for a detailed weighting of different outcomes, optimizing the overall success rate. “Hence, we bypass the restriction by accepting unpredictability, which follows the spirit of quantum mechanics,” the paper states, highlighting the intentional embrace of randomness as a means to achieve deterministic results.

While currently focused on pure states due to their suitability for Gram matrix description, the researchers acknowledge that extending the method to mixed states requires further investigation, though the underlying principle of using multiple Kraus map outcomes remains relevant. The method’s potential extends to one-state exclusion, an important aspect of defining the ontological properties of quantum states, though the output matrices differ in form from those typically used in this context.

Postselection and Unpredictability Bypass Quantum Restrictions

The ability to discriminate between quantum states reaches a new threshold with a method achieving success probabilities of one, even when dealing with states that are not orthogonal and linearly dependent. This bypasses a fundamental limitation of quantum mechanics, which typically restricts such probabilities to values below one, and introduces a level of certainty previously unattainable in these scenarios. The approach relies on accepting unpredictability in the outcome, a strategy that aligns with established quantum principles while circumventing conventional restrictions.

This success isn’t simply a theoretical possibility. The method extends to practical applications in quantum cryptography, specifically in basis reconciliation procedures.

The technique’s efficacy is demonstrated through the error-free discrimination of a single state from a subset of others, a process akin to quantum state filtering. This discrimination is achieved because the resulting output states are orthogonal, allowing for clear identification of the subset through measurement. The method extends beyond filtering a single state, enabling the filtering of multiple states within the set without encountering failure.

This suggests a broader applicability to quantum state exclusion tasks, where the goal is to eliminate specific states from a larger ensemble. The researchers note that this approach allows for obtaining error-free information with a probability exceeding the probability of detecting a single photon, which demonstrates its efficiency and potential.

Non-Orthogonal State Filtering via Division Decomposition

The conventional limit of quantum subsets discrimination, a probability always below one for non-orthogonal states, is circumvented by a new method achieving error-free discrimination through unpredictable subsets division. This approach demonstrates that the inherent uncertainty of quantum mechanics does not invariably restrict operational success, a finding detailed in recent work. The method relies on a decomposition allowing for a success probability of one, even when input states are both non-orthogonal and linearly dependent.

This advancement moves beyond maximizing success for a single division, instead focusing on incorporating multiple, randomized divisions to achieve overall error-free filtering. The researchers detail a construction involving the Gram matrix of three non-orthogonal states, mapping them onto three states where two coincide, resulting in successful filtering for any given division. While individual state discrimination achieves a probability of 1/3, the method’s strength lies in considering several divisions simultaneously, effectively bypassing the previously established limitations.

A general decomposition into a convex combination allows for inclusion of a higher number of subsets for non-binary discrimination, such as a division into N parts corresponding to USD. Average success probability is calculated based on a priori state probabilities and the set of conclusive results, demonstrating the method’s adaptability. The implications extend to quantum cryptography, offering a path to unit sifting probability in basis reconciliation.

Quantum Information Recycling and Randomized Measurement

Error-free discrimination between a quantum state and a subset of other states is now possible, even when those states are not orthogonal, through a technique resembling quantum information recycling. This approach allows for subsequent filtering attempts based on the decomposition of the failure Gram matrix, effectively repurposing information that would otherwise be lost. The process doesn’t simply maximize success for a single division of states, but incorporates multiple possibilities, enhancing overall performance.

This recycling of quantum information is achieved by strategically sacrificing success probability for one particular partitioning of states, allowing for additional filtering attempts on other partitions. The method’s effectiveness hinges on the absence of failure in certain operations, meaning that when a state is identified as belonging to a subset, the identification is guaranteed. This contrasts with traditional quantum state discrimination, where non-orthogonality introduces inherent uncertainty and limits the probability of correct identification.

The implications for quantum cryptography are significant. The proposed method allows for a secret key generation process with a sifting probability of one. This means a large number of bases, even those containing non-orthogonal states, can still achieve unit sifting probability, potentially offering an advantage over existing key sifting methods.

👉 More information
🗞 Quantum subsets discrimination and key sifting with success probability one
✍️ D. A. Kronberg
🧠 ArXiv: https://arxiv.org/abs/2611.09604v1

More like this