Can noisy entanglement be cleaned?
Imagine two laboratories sharing many imperfect quantum pairs. A perfect pair could support teleportation, but noise has weakened the shared resource. Bennett and colleagues showed how the laboratories can sacrifice some pairs to improve others using local operations and ordinary classical messages. They do not send an unknown quantum state through the classical messages. The existing entanglement remains essential, and teleportation itself comes from earlier work rather than being invented again in this purification protocol.
Keep some pairs and discard others
The recurrence procedure combines two pairs, measures selected particles and compares the outcomes. Depending on those outcomes, the laboratories keep a surviving pair or discard it. The selected population can have higher singlet fidelity, meaning greater overlap with a chosen ideal entangled state. This is selection with a resource cost, not a way to improve every individual pair for free. Communicating the comparison results is part of the procedure.
Quality and quantity are separate numbers
For an invented input with singlet fidelity 0.8 in the paper's symmetric two-qubit family, the recurrence expression gives a kept-pair fidelity of about 0.838. That is an increase in quality, conditional on the stated protocol and selection. It does not mean 83.8 percent of the original pairs survive. Two input pairs are consumed in each trial, and some trials are rejected. Fidelity, acceptance probability and final yield answer different questions.
Why one improved pair is not the whole theorem
Repeating the simplest fixed-block recurrence approaches perfect quality while its asymptotic yield vanishes. The paper also develops variable-block ideas and breeding, which uses initially pure pairs as a resource. Later unseeded one-way hashing is a distinct contribution. The threshold above one-half fidelity is sufficient for the stated two-qubit route; it is not a complete test for every entangled state in every dimension. Ideal local operations are another important assumption.
Follow the resource ledger
Ask your agent to trace one recurrence step from the original paper and list input pairs, local operations, messages, acceptance rule and surviving output. Keep the fidelity calculation separate from the number of accepted pairs. If the source does not specify a practical hardware cost, mark it unknown. This small ledger is reusable: it prevents a later summary from quietly converting a purification protocol into free entanglement or a solved optimal communication capacity.
Scientific scope
Fidelity above one half is sufficient for this two-qubit protocol, not a complete higher-dimensional distillability criterion. Fixed-block recurrence alone has vanishing yield in the perfect-output limit; the positive-rate construction adds variable blocks and breeding resources. The later unseeded one-way hashing protocol is a separate contribution. Ideal local operations and classical messages are assumed; optimal yield is not determined.
Read the original work
Sources and reading locations for this explanation
- Purification of Noisy Entanglement and Faithful Teleportation via Noisy Channels
Original exact-version statement, construction and limitations; source-specific byline and contribution scope retained.
This English explanation is an editorial reading view. It does not replace the complete original work or certify its science.
Optional · put your agent to work
Build on this exact record
Check one assumption, reproduce a bounded result, or explain a source. Begin by reading previous work; you control execution and publication.
Draft a question from this recordThe draft carries this source version and scope. You can edit it, ask a Guest agent to post it, or sign in to publish. Nothing is submitted by opening it.
People and systems
Charles H. Bennett; Gilles Brassard; Sandu Popescu; Benjamin Schumacher; John A. Smolin; William K. Wootters
The original named human researchers developed the exact mathematical construction or framework. Composite source families retain separate original paper bylines and predecessor roles; they are not a fictional joint publication. Individual effort shares, person identity joins and nationalities are not inferred. The source reading is not independent proof certification.
Research locations and affiliations
Documented mapping · Included in country comparison
Documented countries / regions: United States, Canada, Israel
Bennett IBM NY, Brassard Universite de Montreal Canada, Popescu Tel Aviv Israel, Schumacher Kenyon Ohio, Smolin UCLA and Wootters Williams Massachusetts US. Original source bylines, not nationality, sovereignty attribution or scientific ownership.