Imagine two physicists, Alice and Bob, located in separate laboratories. They are handed a shared quantum system and tasked with a simple game: determine exactly which state, out of a known set of possibilities, they were given.
If Alice and Bob could bring their particles together into a single room, the task would be trivial. By performing a joint, global measurement across the entire system, they could perfectly distinguish any set of orthogonal quantum states. However, they are physically separated and restricted to Local Operations and Classical Communication (LOCC). They can manipulate their own local subsystems as intricately as they like, and they can chat over a classical telephone line, but they cannot transmit quantum bits or perform joint quantum measurements.
This locality constraint introduces a profound gap between what is globally visible and what is locally accessible. Intuitively, one might assume this gap only matters when the states are entangled. After all, entanglement is the quintessential global quantum resource. Yet, a remarkable body of research reveals that this limitation runs much deeper: even completely unentangled product states can be impossible to tell apart locally. This phenomenon is known as nonlocality without entanglement.
The Intuition Behind Nonlocality Without Entanglement
Why should states that possess absolutely no entanglement be locally indistinguishable? The core of the problem lies in how measurement collapses quantum information.
When Alice performs a local measurement on her subsystem, she extracts some information, but she also forces her part of the quantum state to collapse. If the identity of the global state is hidden within the specific coordination or structure of how Alice's and Bob's states are arranged relative to one another, Alice's local collapse can irreversibly destroy the very correlations Bob needs to observe. Even though the states are mathematically orthogonal and contain no entanglement, the act of looking at them locally erases the global context.
Chitambar et al. provided a rigorous proof of this phenomenon by analyzing the "double trine" ensemble—a specific set of unentangled two-qubit states. The authors demonstrated a strict hierarchy of power, proving that global measurements achieve a fundamentally higher success probability than LOCC protocols, even when the states lack entanglement entirely. This confirmed that state distinguishability is a collective property that cannot always be recovered through piecemeal local pieces.
Global Measurement: [ Alice + Bob ] ---> Perfect Distinction (100%)
LOCC Constraint: [ Alice ] --(Classical Phone)--> [ Bob ] ---> Information Loss
To understand the boundaries of this limitation, Duan et al. established a universal lower bound on the number of states in a multipartite basis that can be unambiguously identified under LOCC. By constructing a "worst-case" basis, they proved that this bound is perfectly tight, demonstrating that local constraints impose hard mathematical limits on information retrieval. Furthermore, Yuan et al. demonstrated "strong" quantum nonlocality without entanglement, showing that entire sets of orthogonal product states can completely resist local discrimination, cementing the idea that nonlocality and entanglement are distinct quantum resources that can exist independently.
The Many-Copy Regime and Mathematical Relaxations
If a single copy of a state cannot be identified locally, a natural question arises: does the problem become easier if Alice and Bob are given multiple copies of the identical state?
Winter et al. investigated this many-copy setting and showed that while the probability of making an error decays exponentially as the number of copies increases, the fundamental constraints of LOCC still dictate the exact rate of this decay. Local observers are still fundamentally disadvantaged compared to global ones, even when flooded with data.
Because analyzing LOCC mathematically is notoriously difficult due to its interactive, multi-round nature, researchers frequently study a broader, more mathematically tractable class of operations: Positive Operator-Valued Measures with Positive Partial Transpose (PPT POVMs). Every valid LOCC measurement is guaranteed to be a PPT operation, so if a state cannot be distinguished by PPT operations, it is definitively indistinguishable by LOCC.
Using this framework, Yu et al. proved that certain entangled states and their complements remain completely indistinguishable by PPT POVMs, even when the observers are granted an infinite number of copies. Similarly, Li et al. linked this multi-copy indistinguishability to the existence of specific algebraic structures called "PPT-definite" operators. These results show that the barrier imposed by physical separation is not just a consequence of small sample sizes; some quantum information is permanently locked away from local view.
A Surprising Asymmetry: Distinguishing Operations
Given that quantum states are so difficult to tell apart locally, one might expect that distinguishing between distributed quantum operations—such as an unknown shared quantum gate or unitary channel—would be equally restricted. Surprisingly, the physics behaves entirely differently.
In a foundational pair of papers, Runyao Duan proved that any two distinct multipartite unitary operations can be perfectly distinguished using only LOCC, provided they can be applied a finite number of times. This means that the nonlocal identity of a quantum gate—even one that creates entanglement—can be fully recovered by distant parties without needing any pre-shared entanglement or joint quantum measurements.
Runyao Duan later formalized the exact mathematical criteria and optimal protocols for this task in /paper/art_0c9240a6bbb44427b53b067e00ec7c24, proving that a finite number of queries is always sufficient to achieve 100% distinguishability.
Why does this asymmetry exist? When distinguishing states, Alice and Bob are passive observers trying to read static information that can be easily destroyed by collapse. When distinguishing operations, they can be active participants. They can feed a carefully chosen input state into the unknown operation, pass the resulting state through a sequence of local steps, and reuse the operation multiple times. This sequential feedback loop allows the global parameters of the operation to gradually imprint themselves onto the local subsystems, eventually making them perfectly visible to LOCC.
Bridging the Local-Global Gap
Understanding these limits allows researchers to quantify exactly what resources are needed to overcome them. If Alice and Bob want to make their local measurements just as powerful as global ones, they must pay a price in other quantum currencies:
- Entanglement Cost: Zhu et al. quantified the exact amount of pre-shared entanglement required to bridge the local-global gap. They proved that while a single Bell state (1 ebit) can always unlock the hidden information in a pure state, mixed states are far more stubborn, requiring significantly higher amounts of entanglement to be fully identified.
- Communication Constraints: If the classical communication link itself is restricted—such as the "zero-rate" regime where the amount of data Alice and Bob can exchange vanishes asymptotically—Sreekumar et al. showed that distributed hypothesis testing becomes a delicate trade-off between local data reduction and the preservation of global correlation information.
These fundamental limits also carve out boundaries for quantum cryptography. For example, Ananth et al. explored the Common Haar State model, where distributed parties share random quantum states. They proved that an LOCC adversary's advantage in distinguishing identical versus independent random quantum states is strictly bounded by $O(t^2/2^n)$ (where $t$ is the number of copies and $n$ is the number of qubits). This mathematical bound ensures that shared quantum randomness remains securely hidden from observers who cannot perform joint quantum measurements.
Ultimately, as Lu et al. demonstrated through a rigorous no-go theorem for many-body systems, the space of reachable quantum observables is severely restricted by the locality of physical interactions. The fact that information can be distributed nonlocally without entanglement is not just a mathematical curiosity; it is a fundamental feature of quantum mechanics that defines how, and to what extent, we can learn about a distributed quantum world.
Key Foundational Work
Distinguishing Arbitrary Multipartite Basis Unambiguously Using Local Operations and Classical Communication (2007) — contributed 2 concept elements to this direction.
Local Distinguishability of Multipartite Unitary Operations (2008) — contributed 2 concept elements to this direction.