Post-Quantum Signatureless Authorization
Year: 2026
Type: preprint
Abstract: Digital signatures authorize transactions, commands and other requests. Quantum attacks threaten widely used signature schemes, while their post-quantum replacements can be costly to verify inside proofs of execution. Signatureless authorization binds knowledge of an authority secret to a canonical request through a freshly randomized proof capsule. The public statement consists of a request identifier and a 32-byte authority commitment. Application rules enforce context, policy and freshness. The interface supports standalone requests and recursive proof composition. The construction combines a masked degree-ten authority reduction over GF(2^128), embedded in GF(2^256), with a bank-and-companion code proof. Source restoration gives expected polynomial-time public-coin extraction. An adaptive witness-free simulator and a same-experiment source extractor establish a quantum chosen-request authorization reduction in the stated address-compatible random-completion model. The public authority circuit and its address input slice are fixed. One permutation supplies the native transcript and commitments. Finite orbit-profile and resource conditions are explicit. The reduction quantifies simulation and extraction losses alongside authority-preimage hardness and canonical intent binding. For S <= T and T <= 2^24 charged native calls, the combined simulation and extraction loss is below 2^-67, including the uniform-slice setup failure. Parano1d supplies a blockchain implementation and recursive verifier. Canonical capsules are bounded by 92,696 bytes. On four CPU threads, median proving takes 0.81 to 0.89 seconds and verification takes approximately 25.6 milliseconds for two request fixtures. The accompanying reproducibility archive contains the manuscript sources, exact arithmetic and finite operator checks, simulation tests, recorded measurements, Rust reproduction programs and authenticated production source files. Its README provides the reproduction commands. Contact: [email protected] information: The paper and accompanying written documentation are licensed under Creative Commons Attribution 4.0 International (CC BY 4.0). Original software in the reproducibility artifact is licensed under the Apache License, Version 2.0. Third-party dependencies retain their respective licenses.
Author: Nemo, Ignotus
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7224343149