Variations of the primitive normal basis theorem: exact Gauss-sum identities, two uncovered pairs, and a sieve proof in characteristic 3

This record contains a note, with its verification program and data, examining the proofs in G. Kapetanakis and L. Reis, "Variations of the primitive normal basis theorem", Des. Codes Cryptogr. 87 (2019), 1459-1480, doi:10.1007/s10623-018-0543-9. That paper studies elements of F_{q^n} of multiplicative order (q^n-1)/2 that are normal or 1-normal over F_q. Two texts of the authors were compared: the arXiv version (arXiv:1712.09861v1) and the revised manuscript dated August 5, 2018. The printed article was not available to us. Findings that hold for both texts:- The identities of Theorem 3.1 are false: a family of quadratic Gauss sums was set to zero. The inequality drawn from them remains valid; the exact identity is given (Theorem C).- The lemma on n = 2 is proved from a false inequality; a short repair is given.- The pairs (q,n) = (5,10) and (7,14) are not covered: the decisive inequality fails there although it is stated to have been verified. Explicit 2-primitive 1-normal elements are given for both. Points specific to one text: the arXiv version misstates the case n = 2, omits a factor 2 for n = p and n = 2p, and excludes characteristic 3. The revised manuscript corrects these and covers characteristic 3; as written, its condition for n = p fails at (7,7), which is recovered with the exact count of primitive square roots. For characteristic 3 the note gives a second, independent proof (Theorem B): for every q = 3^t and every n >= 3 there is an element of order (q^n-1)/2 with F_q-order (x^n-1)/(x-1), hence 1-normal over F_q. The statement is already contained in the revised manuscript of Kapetanakis and Reis; the proof here is different. It uses a prime sieve, four criteria, a finite certification of 4002 pairs (t,n), and 22 explicit elements, and it does not rely on the faulty identity. Contents:- kapetanakis_reis_pnbt_variations_audit.pdf / .tex: the note (8 pages).- code/verify_kapetanakis_reis_dcc2019.py: deterministic Python verification program (requires sympy; matplotlib for the figures).- data/: program output (JSON files and the run log), including two independent checks of each of the 22 explicit elements.- figures/: the two figures of the note. Scope: part (1) of the main theorem for n >= 3 and part (2) for p >= 5 rest on the computations and tables of Kapetanakis and Reis, which were not re-examined. AI disclosure: Claude (Anthropic) and Gemini (Google) were used to assist with the verification scripts, the checking of identities, and the drafting of the manuscript; the author reviewed the content and takes full responsibility for it.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-20
DOI
https://doi.org/10.5281/zenodo.24397996
Primary Topic
Coding theory and cryptography
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Variations of the primitive normal basis theorem: exact Gauss-sum identities, two uncovered pairs, and a sieve proof in characteristic 3

Yuqing Cheng
1 links
Zenodo (CERN European Organization for Nuclear Research)
Coding theory and cryptography
preprint

Variations of the primitive normal basis theorem: exact Gauss-sum identities, two uncovered pairs, and a sieve proof in characteristic 3

Yuqing Cheng
preprint en

Abstract

This record contains a note, with its verification program and data, examining the proofs in G. Kapetanakis and L. Reis, "Variations of the primitive normal basis theorem", Des. Codes Cryptogr. 87 (2019), 1459-1480, doi:10.1007/s10623-018-0543-9. That paper studies elements of F_{q^n} of multiplicative order (q^n-1)/2 that are normal or 1-normal over F_q. Two texts of the authors were compared: the arXiv version (arXiv:1712.09861v1) and the revised manuscript dated August 5, 2018. The printed article was not available to us. Findings that hold for both texts:- The identities of Theorem 3.1 are false: a family of quadratic Gauss sums was set to zero. The inequality drawn from them remains valid; the exact identity is given (Theorem C).- The lemma on n = 2 is proved from a false inequality; a short repair is given.- The pairs (q,n) = (5,10) and (7,14) are not covered: the decisive inequality fails there although it is stated to have been verified. Explicit 2-primitive 1-normal elements are given for both. Points specific to one text: the arXiv version misstates the case n = 2, omits a factor 2 for n = p and n = 2p, and excludes characteristic 3. The revised manuscript corrects these and covers characteristic 3; as written, its condition for n = p fails at (7,7), which is recovered with the exact count of primitive square roots. For characteristic 3 the note gives a second, independent proof (Theorem B): for every q = 3^t and every n >= 3 there is an element of order (q^n-1)/2 with F_q-order (x^n-1)/(x-1), hence 1-normal over F_q. The statement is already contained in the revised manuscript of Kapetanakis and Reis; the proof here is different. It uses a prime sieve, four criteria, a finite certification of 4002 pairs (t,n), and 22 explicit elements, and it does not rely on the faulty identity. Contents:- kapetanakis_reis_pnbt_variations_audit.pdf / .tex: the note (8 pages).- code/verify_kapetanakis_reis_dcc2019.py: deterministic Python verification program (requires sympy; matplotlib for the figures).- data/: program output (JSON files and the run log), including two independent checks of each of the 22 explicit elements.- figures/: the two figures of the note. Scope: part (1) of the main theorem for n >= 3 and part (2) for p >= 5 rest on the computations and tables of Kapetanakis and Reis, which were not re-examined. AI disclosure: Claude (Anthropic) and Gemini (Google) were used to assist with the verification scripts, the checking of identities, and the drafting of the manuscript; the author reviewed the content and takes full responsibility for it.

Zenodo (CERN European Organization for Nuclear Research)
Coding theory and cryptography
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon