Karim Eldefrawy

Cryptography, Cybersecurity, Privacy

Co-founder and CTO at Confidencial.io
2017-2021: SRI
2011-2016: HRL Laboratories
2006-2010: PhD@UC Irvine

Scientific curiosity

Scientific knowledge map · Paper #56

Optimizing Registration Based Encryption

Kelong Cong, Karim Eldefrawy, and Nigel P. Smart

2021 · 18th IMA International Conference on Cryptography and Coding (IMACC)

  • Theory
  • Applied
  • scheme

What does the paper try to establish?

Can registration-based encryption preserve its no-key-escrow and asymptotic efficiency goals while reducing the enormous concrete ciphertext and decryption costs of existing constructions?

What is the proposed answer?

Replacing Merkle trees with authenticated crit-bit trees shortens garbled inputs, and publishing a small square-root-sized identity filter lets encryptors skip roughly half the tree snapshots. The paper analytically estimates a 57.5 percent combined ciphertext reduction, proves standard RBE security for the crit-bit construction, and states important remaining security and practicality gaps.

Six dimensions, kept separate

The chart summarizes documented evidence and process. It is not a correctness probability, confidence score, or ranking, and no composite score is calculated.

The visual spider chart requires JavaScript. The complete values and rationales follow in text.

LowMediumHighN/A = not assessed

A smaller value means less documented support for that dimension, not that the paper is false or unimportant.

Epistemic evidence Medium

The construction and standard-security proof are detailed, but the paper's central concrete-efficiency claims are analytical estimates without an implementation, and verifiability is only sketched.

Single-user and general RBE security proofs Analytical cost reduction Verifiability argument and proof boundary CCA, implementation, and compactness boundaries
Auditability High

Full archive and author-uploaded copies plus the DOI expose the definitions, proofs, and arithmetic. Anti-bot controls prevented recording a local byte hash in this pass.

Author-uploaded full text Official IMACC publication identity
Production provenance Medium

Authorship, venue, DOI, archive, and author upload are documented; revision history, roles, and tool use are not.

Official IMACC publication identity RBE motivation, baseline cost, and contributions
External scrutiny Medium

IMACC publication establishes venue review, but no public review report, artifact evaluation, or independent proof audit is represented.

Official IMACC publication identity
Reception Low

The dated exact-DOI OpenAlex record located 0 citations. Under the author-defined rule, 0 through 8 located citations is Low; counts vary across indexes and dates.

Dated citation-count snapshot
Contribution significance Medium

The work materially reduces an identified concrete bottleneck in the prior blueprint, but the resulting construction remains impractical and downstream impact is not established here.

RBE motivation, baseline cost, and contributions CCA, implementation, and compactness boundaries

Assessment: Ai draft author review pending · 2026-07-11 · rubric 0.2. These dimensions describe documented support and process, not truth, correctness, or a universal ranking. No composite score is calculated.

Hierarchical knowledge map

Collapse a branch for a top-level reading, or follow its source links and child nodes to audit the evidence and boundaries underneath it.

paper

Optimizing Registration Based Encryption

An optimized RBE scheme using authenticated crit-bit trees and a public-parameter tradeoff to reduce the concrete cost of garbled public-key operations.

RBE motivation, baseline cost, and contributions
  1. scope RBE model defined

    Users generate their own keys; a key curator holds no secret and maintains registrations, auxiliary state, and short public parameters; an encryptor uses recipient identity plus public parameters, and a decryptor combines its secret key with periodically updated supporting information.

    RBE roles, algorithms, efficiency, and message-hiding definition
  2. scheme Crit-bit-tree RBE specified

    Authenticated crit-bit trees replace balanced Merkle trees in snapshot-based registration. Each internal node stores a critical bit and two hash pointers, shortening the garbled input while preserving searchable authenticated paths.

    Authenticated crit-bit tree Compact-parameter optimized RBE construction
    1. claim

      Compact-parameter reduction

      analytical estimate

      For the paper's two-billion-user, 256-bit example, shorter tree nodes are estimated to reduce encryptor public-key operations and ciphertext size by about 15 percent and decryptor work by about 30 percent on average.

      Analytical cost reduction
  3. method Larger public parameter specified tradeoff

    Publishing membership information for a square-root-sized minority of snapshot users lets encryption ignore roughly half the trees. A cuckoo filter is proposed to represent that set compactly.

    Larger-public-parameter optimization
  4. claim

    Verifiable registration compatibility

    argued not fully proved

    Adjacent authenticated paths support non-membership and unique-membership proof sketches, but the paper explicitly does not provide the full verification algorithm or prove its soundness and completeness.

    Verifiability argument and proof boundary
  5. scrutiny

    External scrutiny

    venue reviewed

    The paper appeared at IMACC 2021. Review reports and independent implementation or proof audits are not linked.

    Official IMACC publication identity

Source index

Locators state the depth of the current audit. PDF page numbers, where present, are one-based file pages; metadata-, summary-, and abstract-bounded records explicitly identify their limitations.

  1. RBE motivation, baseline cost, and contributions Abstract and Sections 1.1-1.2, PDF pages 1-3
  2. Author-uploaded full text ResearchGate public full-text record uploaded by Karim Eldefrawy on 2021-09-29
  3. RBE roles, algorithms, efficiency, and message-hiding definition Sections 2 and 3.5, PDF pages 4-12
  4. Hash garbling, hash encryption, and cryptographic assumptions Sections 3.3-3.4 and Appendix A, PDF pages 9-10 and 25-28
  5. Authenticated crit-bit tree Sections 3.6 and 4.1, PDF pages 12-14
  6. Compact-parameter optimized RBE construction Sections 4.2-4.3, PDF pages 14-17
  7. Single-user and general RBE security proofs Section 4.4, Theorems 1-2, PDF pages 17-19
  8. Analytical cost reduction Section 4.5, PDF page 19
  9. Verifiability argument and proof boundary Section 4.6, PDF pages 19-20
  10. Larger-public-parameter optimization Section 5, PDF pages 20-23
  11. CCA, implementation, and compactness boundaries Sections 1.2, 4.5-4.6, and 5, PDF pages 3 and 19-23
  12. Official IMACC publication identity DOI 10.1007/978-3-030-92641-0_7
  13. Dated citation-count snapshot OpenAlex reported 0 citations when accessed 2026-07-11