Linear Codes Over Finite Fields From Lattice Polytopes: Structure, Duality, and Quantum Constructions
Sihem Mesnager, Rameez Raja
Linear codes constructed from combinatorial and geometric structures have played a central role in both classical and quantum coding theory. In this paper, we discuss a class of linear codes over finite fields, calledpolytope codes, obtained via defining sets derived from lattice polytopes. Unlike evaluation-based constructions, our approach exploits the intersection structure of polytope faces to directly control the algebraic and metric properties of the resulting codes. We develop a general defining-set framework for polytope codes and explicitly determine their lengths, dimensions, minimum distances, and Hamming weight distributions. This leads to infinite families of projective, few-weight, distance-optimal, and almost distance-optimal codes. Sufficient conditions for minimality are derived, resulting in new minimal codes that are well suited for applications to secret-sharing schemes. We further study the duals and hulls of polytope codes and provide explicit characterizations of self-orthogonal and linear complementary dual (LCD) families. One of the central results of the paper is a complete classification of quantum codes arising from polytope codes via the Calderbank-Shor-Steane (CSS) construction. We show that the geometry of face intersections fully determines whether the resulting quantum code is a standard CSS code, an LCD-based code, or an entanglement-assisted quantum error-correcting code, and we explicitly compute the required entanglement in the latter case. As a consequence, new families of binary, ternary, and prime-power quantum codes with previously unattained parameters are obtained. The proposed polytope-based framework unifies classical and quantum code constructions under a common geometric principle and opens new avenues for applications in cryptography, distributed storage, and quantum information processing.Mathematics Subject Classification. 94B05, 52B20, 11T71, 94A62, 94A45.
IEEE Transactions on Information Theory•Sep 7, 2026•0 citations