{"schema_version":"0.1","map_id":"paper-19-map","publication_id":19,"publication_anchor":"paper-19","slug":"paper-19","canonical_path":"/knowledge/papers/paper-19/","machine_path":"/knowledge/papers/paper-19.json","root_node_id":"paper-19","stage":"mapped_draft","contribution_type_vocabulary_version":"0.1","contribution_types":["protocol","primitive"],"title":"5PM: Secure Pattern Matching","year":2012,"venue":"8th Conference on Security and Cryptography for Networks (SCN)","topic":"secure-encrypted-computation","labels":["Theory"],"authors":["Joshua Baron","Karim Eldefrawy","Kirill Minkovich","Rafail Ostrovsky","Eric Tressler"],"keywords":["secure pattern matching","homomorphic encryption"],"research_question":"Can two parties privately evaluate expressive pattern queries—exact match, single-character wildcards, non-binary substring/Hamming matching, and optionally hidden pattern length—with communication below the size of a generic matching circuit and security against malicious participants?","central_answer":"5PM turns pattern matching into linear operations over character-delay vectors and an activation vector, enabling an honest-but-curious protocol from generic additively homomorphic encryption and a malicious protocol from threshold ElGamal plus zero-knowledge proofs. The source proves conditional security and gives sublinear-in-circuit communication and low round counts; only the honest-but-curious variant is implemented and benchmarked in the audited full version.","curation":{"drafted_at":"2026-07-11","drafted_by":[{"actor_type":"ai","name":"OpenAI Codex","role":"full-text extraction, protocol and proof decomposition, evidence linking, and initial assessment"}],"method":"Source-grounded review of the complete public IACR ePrint full version, the SCN publisher record, an author-uploaded full-text route, and a dated ResearchGate citation snapshot. Proof statements and benchmarks were inspected but not independently re-proved or rerun.","source_scope":"full_source_audit","approval":{"status":"pending","note":"AI-authored source-linked map awaiting author verification; technical summaries and ratings may be revised before approval."}},"sources":[{"id":"source-paper-19-eprint","type":"scholarly_preprint","title":"5PM: Secure Pattern Matching, full version","url":"https://eprint.iacr.org/2012/698.pdf","media_type":"application/pdf","page_count":51,"provenance_category":"public_archive","retrieved_at":"2026-07-11"},{"id":"source-paper-19-official","type":"publication_record","title":"SCN 2012 publisher record","url":"https://doi.org/10.1007/978-3-642-32928-9_13"},{"id":"source-paper-19-author","type":"author_hosted_copy","title":"ResearchGate author-uploaded full text","url":"https://www.researchgate.net/publication/266956911_5PM_Secure_Pattern_Matching","provenance_category":"author","retrieved_at":"2026-07-11"},{"id":"source-paper-19-journal","type":"subsequent_publication_record","title":"Journal of Computer Security full-version record","url":"https://journals.sagepub.com/doi/10.3233/JCS-130481","year":2013},{"id":"source-paper-19-citations","type":"citation_index_snapshot","title":"ResearchGate citation snapshot","url":"https://www.researchgate.net/publication/266956911_5PM_Secure_Pattern_Matching","retrieved_at":"2026-07-11"}],"source_anchors":[{"id":"anchor-paper-19-question","source_id":"source-paper-19-eprint","label":"Problem, functionality, and claimed contributions","locator":"Abstract and Section 1, PDF pages 1-4","url":"https://eprint.iacr.org/2012/698.pdf#page=1"},{"id":"anchor-paper-19-ipm","source_id":"source-paper-19-eprint","label":"Insecure pattern-matching algorithm and character-delay vectors","locator":"Section 2.1, PDF pages 5-7","url":"https://eprint.iacr.org/2012/698.pdf#page=5"},{"id":"anchor-paper-19-linear","source_id":"source-paper-19-eprint","label":"Linear-algebra formulation and Stretch/Cut/ColSum operations","locator":"Sections 2.2-2.3, PDF pages 7-10","url":"https://eprint.iacr.org/2012/698.pdf#page=7"},{"id":"anchor-paper-19-hbc","source_id":"source-paper-19-eprint","label":"Honest-but-curious 5PM protocol","locator":"Section 3.2, PDF pages 11-12","url":"https://eprint.iacr.org/2012/698.pdf#page=11"},{"id":"anchor-paper-19-malicious","source_id":"source-paper-19-eprint","label":"Malicious-static-corruption protocol and zero-knowledge checks","locator":"Section 3.3, PDF pages 12-16","url":"https://eprint.iacr.org/2012/698.pdf#page=12"},{"id":"anchor-paper-19-theorems","source_id":"source-paper-19-eprint","label":"Security theorems and complexity claims","locator":"Theorems 1-2 and Section 4, PDF pages 12-18","url":"https://eprint.iacr.org/2012/698.pdf#page=12"},{"id":"anchor-paper-19-proofs","source_id":"source-paper-19-eprint","label":"Simulation-based proof details","locator":"Proof sections and appendices, PDF pages 32-48","url":"https://eprint.iacr.org/2012/698.pdf#page=32"},{"id":"anchor-paper-19-implementation","source_id":"source-paper-19-eprint","label":"Honest-but-curious implementation and benchmark tables","locator":"Implementation section and Table 13, PDF pages 48-50","url":"https://eprint.iacr.org/2012/698.pdf#page=48"},{"id":"anchor-paper-19-publication","source_id":"source-paper-19-official","label":"Official conference publication record","locator":"SCN 2012, pages 222-240, DOI record","url":"https://doi.org/10.1007/978-3-642-32928-9_13"},{"id":"anchor-paper-19-journal","source_id":"source-paper-19-journal","label":"Subsequent journal full-version record","locator":"Journal of Computer Security 21(5), pages 601-625","url":"https://journals.sagepub.com/doi/10.3233/JCS-130481"},{"id":"anchor-paper-19-citations","source_id":"source-paper-19-citations","label":"Citation-count snapshot","locator":"ResearchGate displayed Citations (56), observed 2026-07-11; coverage and merging of conference, ePrint, and journal versions may differ from other indexes.","url":"https://www.researchgate.net/publication/266956911_5PM_Secure_Pattern_Matching"}],"nodes":[{"id":"paper-19","kind":"paper","parent_id":null,"order":1,"epistemic_status":"published","title":"5PM","summary":"A two-party secure pattern-matching protocol family based on an algebraic reduction and additively homomorphic encryption.","source_anchor_ids":["anchor-paper-19-question"]},{"id":"paper-19-question","kind":"question","parent_id":"paper-19","order":1,"epistemic_status":"research_question","title":"Research question","summary":"Can expressive non-binary pattern search be securely computed with communication linear in text-plus-pattern length rather than linear in a quadratic generic circuit?","source_anchor_ids":["anchor-paper-19-question"]},{"id":"paper-19-answer","kind":"contribution","parent_id":"paper-19","order":2,"epistemic_status":"formally_supported","title":"Central answer","summary":"Represent candidate matches as linear matrix/vector operations over character-delay vectors, then evaluate those operations homomorphically with protocol-specific checks for semi-honest or malicious behavior.","source_anchor_ids":["anchor-paper-19-linear","anchor-paper-19-hbc","anchor-paper-19-malicious"]},{"id":"paper-19-scope","kind":"scope","parent_id":"paper-19","order":3,"epistemic_status":"explicitly_scoped","title":"Functionality and parties","summary":"A server holds text of length n and a client holds a pattern of length m over an arbitrary alphabet; outputs can represent exact, single-wildcard, substring/Hamming matches, with optional pattern-length hiding.","source_anchor_ids":["anchor-paper-19-question"]},{"id":"paper-19-adversary","kind":"threat_model","parent_id":"paper-19-scope","order":1,"epistemic_status":"defined","title":"Semi-honest and malicious models","summary":"Separate constructions address honest-but-curious behavior and static malicious corruption in the stand-alone model; privacy is defined through simulation and allowed leakage such as public length bounds.","source_anchor_ids":["anchor-paper-19-hbc","anchor-paper-19-malicious","anchor-paper-19-proofs"]},{"id":"paper-19-primitive-ipm","kind":"primitive","parent_id":"paper-19","order":4,"epistemic_status":"algorithmically_defined","title":"Character-delay-vector matching","summary":"Each pattern character creates delayed contributions to an activation vector; a match is detected where the accumulated value equals the pattern length, with adjustments for wildcards and allowed Hamming distance.","source_anchor_ids":["anchor-paper-19-ipm"]},{"id":"paper-19-primitive-linear","kind":"primitive","parent_id":"paper-19","order":5,"epistemic_status":"proposed","title":"Linear-algebra reduction","summary":"Matrix multiplication followed by Stretch, Cut, and column summation expresses matching using additions and scalar products compatible with additive homomorphism.","source_anchor_ids":["anchor-paper-19-linear"]},{"id":"paper-19-protocol-hbc","kind":"protocol","parent_id":"paper-19","order":6,"epistemic_status":"proven_conditional_and_implemented","title":"Two-round honest-but-curious protocol","summary":"The client encrypts preprocessed pattern material, the server evaluates the linear matching computation over ciphertexts, and the client decrypts the blinded result; security relies on semantic security of the chosen additive homomorphic scheme.","source_anchor_ids":["anchor-paper-19-hbc","anchor-paper-19-theorems"]},{"id":"paper-19-protocol-malicious","kind":"protocol","parent_id":"paper-19","order":7,"epistemic_status":"proven_conditional_not_implemented","title":"Eight-round malicious protocol","summary":"Threshold ElGamal, commitments, and interleaved zero-knowledge arguments constrain both parties to well-formed encrypted inputs and protocol steps while preserving the algebraic matching path.","source_anchor_ids":["anchor-paper-19-malicious","anchor-paper-19-theorems"]},{"id":"paper-19-claim-security-hbc","kind":"formal_claim","parent_id":"paper-19","order":8,"epistemic_status":"theorem_in_source","title":"Semi-honest security","summary":"Theorem 1 establishes simulation-based security of the honest-but-curious construction assuming a semantically secure additively homomorphic encryption scheme over the specified group.","source_anchor_ids":["anchor-paper-19-theorems","anchor-paper-19-proofs"]},{"id":"paper-19-claim-security-malicious","kind":"formal_claim","parent_id":"paper-19","order":9,"epistemic_status":"theorem_in_source","title":"Malicious security","summary":"Theorem 2 establishes security of the malicious construction under the Decisional Diffie-Hellman assumption and the soundness/zero-knowledge properties of its proof subprotocols.","source_anchor_ids":["anchor-paper-19-theorems","anchor-paper-19-proofs"]},{"id":"paper-19-claim-complexity","kind":"formal_claim","parent_id":"paper-19","order":10,"epistemic_status":"asymptotic_analysis","title":"Communication and round complexity","summary":"The malicious construction uses O((m+n)k^2) bandwidth and O(m+n) encryptions and completes in eight rounds; the honest-but-curious construction completes in two rounds, with computational work still scaling with pattern-text interactions.","source_anchor_ids":["anchor-paper-19-question","anchor-paper-19-theorems"]},{"id":"paper-19-evidence-proof","kind":"formal_evidence","parent_id":"paper-19","order":11,"epistemic_status":"human_proof","title":"Simulation proof chain","summary":"The full version supplies hybrid/simulator arguments for the protocol components and their composition; this map records the chain but does not independently verify every proof step.","source_anchor_ids":["anchor-paper-19-proofs"]},{"id":"paper-19-evidence-implementation","kind":"empirical_evidence","parent_id":"paper-19","order":12,"epistemic_status":"measured_prototype","title":"Honest-but-curious prototype","summary":"A C++ Paillier-based implementation on a dual quad-core 2.93 GHz machine reports, for alphabet 36, pattern 100, and text 10,000, about 7.55 seconds at 1024-bit keys and 25.01 seconds at 2048-bit keys.","source_anchor_ids":["anchor-paper-19-implementation"]},{"id":"paper-19-limitations","kind":"limitation_group","parent_id":"paper-19","order":13,"epistemic_status":"material","title":"Implementation and evidence limits","summary":"The malicious protocol is not implemented in the reported evaluation, benchmarks use an older single-machine setup, preprocessing and key generation are separated from online totals, and no public code or independent reproduction was located.","source_anchor_ids":["anchor-paper-19-implementation"]},{"id":"paper-19-artifacts","kind":"artifact_group","parent_id":"paper-19","order":14,"epistemic_status":"paper_and_proof_only","title":"Artifacts","summary":"The IACR full version exposes detailed constructions, proofs, and benchmark tables; an author-uploaded full text is also public, but no code repository, test vectors, or locally fixed PDF was obtained.","source_anchor_ids":["anchor-paper-19-proofs","anchor-paper-19-implementation"]},{"id":"paper-19-scrutiny","kind":"scrutiny","parent_id":"paper-19","order":15,"epistemic_status":"conference_and_journal_reviewed","title":"External scrutiny and reception","summary":"A preliminary version appeared at SCN 2012, a full version later appeared in the Journal of Computer Security, and ResearchGate reports 56 citations; review reports and independent proof/implementation audits were not inspected.","source_anchor_ids":["anchor-paper-19-publication","anchor-paper-19-journal","anchor-paper-19-citations"]},{"id":"paper-19-lineage","kind":"lineage","parent_id":"paper-19","order":16,"epistemic_status":"documented","title":"Research lineage","summary":"5PM contributes a reusable linear representation for secure string search and supplies the cryptographic search primitive later used and surveyed in application-oriented private-search work.","source_anchor_ids":["anchor-paper-19-question","anchor-paper-19-linear"]}],"relations":[{"id":"paper-19-relation-ipm-enables-linear","type":"enables","from_id":"paper-19-primitive-ipm","to_id":"paper-19-primitive-linear"},{"id":"paper-19-relation-linear-enables-hbc","type":"enables","from_id":"paper-19-primitive-linear","to_id":"paper-19-protocol-hbc"},{"id":"paper-19-relation-linear-enables-malicious","type":"enables","from_id":"paper-19-primitive-linear","to_id":"paper-19-protocol-malicious"},{"id":"paper-19-relation-proof-supports-hbc","type":"supports","from_id":"paper-19-evidence-proof","to_id":"paper-19-claim-security-hbc"},{"id":"paper-19-relation-proof-supports-malicious","type":"supports","from_id":"paper-19-evidence-proof","to_id":"paper-19-claim-security-malicious"},{"id":"paper-19-relation-implementation-supports-hbc","type":"supports","from_id":"paper-19-evidence-implementation","to_id":"paper-19-protocol-hbc"},{"id":"paper-19-relation-limitations-qualify-answer","type":"qualifies","from_id":"paper-19-limitations","to_id":"paper-19-answer"}],"assessment":{"id":"paper-19-assessment-2026-07-11","rubric_version":"0.2","assessed_at":"2026-07-11","status":"ai_draft_author_review_pending","note":"These dimensions describe documented support and process, not truth, correctness, or a universal ranking. No composite score is calculated.","axes":[{"id":"epistemic_evidence","level":"high","rationale":"The full version provides explicit functionality and adversary models, theorem statements, extensive simulation-based proofs, asymptotic analysis, and an honest-but-curious implementation with benchmarks. The malicious variant is not implemented and neither proof nor benchmark was independently reproduced in this audit.","basis_source_anchor_ids":["anchor-paper-19-theorems","anchor-paper-19-proofs","anchor-paper-19-implementation"]},{"id":"auditability","level":"high","rationale":"A complete IACR full version and an author-uploaded full text make the construction, proofs, and benchmarks directly inspectable. This audit did not obtain a locally hashed binary or public implementation source.","basis_source_anchor_ids":["anchor-paper-19-question","anchor-paper-19-proofs","anchor-paper-19-implementation"]},{"id":"production_provenance","level":"medium","rationale":"Named authorship, public ePrint, conference DOI, author upload, and later journal record establish a clear publication lineage. Contributor roles, code versions, benchmark scripts, and revision history are not documented.","basis_source_anchor_ids":["anchor-paper-19-question","anchor-paper-19-publication","anchor-paper-19-journal"]},{"id":"external_scrutiny","level":"high","rationale":"The work passed SCN review, received a journal full-version publication, and has a substantial citation trail. Review reports, independent proof checking, and implementation replication were not located.","basis_source_anchor_ids":["anchor-paper-19-publication","anchor-paper-19-journal","anchor-paper-19-citations"]},{"id":"reception","level":"high","rationale":"ResearchGate displayed 56 citations on 2026-07-11, exceeding the rubric's 11-citation high threshold. The count may merge conference, ePrint, and journal versions, and citing contexts were not reviewed.","basis_source_anchor_ids":["anchor-paper-19-citations"]},{"id":"contribution_significance","level":"high","rationale":"The work provides a general algebraic primitive, malicious-security construction, favorable communication/round complexity, and a prototype, and it has a sustained citation and publication trail.","basis_source_anchor_ids":["anchor-paper-19-question","anchor-paper-19-theorems","anchor-paper-19-implementation","anchor-paper-19-citations"]}]},"reception_snapshot":{"as_of":"2026-07-11","method":"researchgate_publication_page","citation_count":56,"source":"ResearchGate publication page","signals":["ResearchGate displayed Citations (56).","A full version subsequently appeared in the Journal of Computer Security."],"limitation":"ResearchGate may merge the conference, IACR ePrint, and journal versions differently from other indexes; citing contexts, polarity, proof reuse, and software adoption were not audited."}}