arXiv:2610.08012
Cyclic triorthogonal codes in prime dimension
Shiroman Prakash
Abstract
We study cyclic triorthogonal codes in prime dimension $p$, on which a transversal third-level diagonal gate acts as a logical non-Clifford gate. For odd $p$, we use a definition of triorthogonality that does not require self-orthogonality. For cyclic codes of length coprime to $p$, triorthogonality reduces to a sumset condition on the spectral support, which makes exhaustive search feasible: for two natural constructions, we enumerate all maximal codes of this kind for $p=3$ up to length $121$ and $p=5,7$ up to length $96$. For qubits, we prove that a cyclic code of odd length is triply even if and only if it is classically triorthogonal, giving a complete description of triply-even cyclic codes, and we determine all such codes up to length $763$. For qudits, we find codes such as $[[19,1,3]]_3$ and $[[7,1,3]]_7$, which improve on the shortest Reed-Muller and Reed-Solomon codes in both overhead exponent and distillation threshold. The latter is the first member of an infinite family of $[[p,1,2\lfloor p/6\rfloor+1]]_p$ codes.
AI Summary
The paper studies cyclic CSS codes supporting transversal third-level diagonal non-Clifford gates, focusing on small prime-dimensional magic-state distillation codes. For lengths coprime to $p$, it expresses triorthogonality through the spectral support $S$: construction A, with cyclic stabilizers and an all-ones logical row, requires $0\notin S\cup2S\cup3S$; construction B punctures a cyclic space satisfying $0\notin3S$. For odd primes, the definition omits self-orthogonality; transversal cubic gates are established directly, with Pauli corrections for qutrits. For binary cyclic codes of odd length, classical triorthogonality is proved equivalent to triply-evenness using McEliece’s weight-divisibility theorem. Maximal supports are exhaustively enumerated through length $763$ for qubits, $121$ for qutrits, and $96$ for $p=5,7$. Search completeness is independently checked, and distances are established by exhaustive search, classical coding bounds, or concatenation, with some remaining lower bounds. The resulting codes include $[[19,1,3]]_3$ and $[[7,1,3]]_7$, with depolarizing-noise distillation thresholds $0.226$ and $0.446$ and overhead exponents $2.68$ and $1.77$, respectively. Spectral constructions and BCH/Singleton bounds also yield an infinite $[[p,1,2\lfloor p/6\rfloor+1]]_p$ family for primes $p\ge7$. Exact single-round performance follows from weight enumerators and MacWilliams identities. The study is restricted to two constructions, unit weights, coprime lengths, and multiplier equivalence. Binary codes do not improve known non-cyclic lengths, and recursive $[[4,1,2]]_5$ distillation remains preferable for $p=5$. Thresholds assume symmetric twirled depolarizing noise; acceptance costs and full multi-round overheads remain unassessed.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 24 pages, 2 tables. Code and data: https://github.com/spdei/cyclic-triorthogonal-codes