KEVOS
ArticlesServicesCase studiesAboutContact
ArticlesServicesCase studiesAboutContact
← ArticlesSplitting Separable Algebras over Finite FieldsEngineering · Engineering MathematicsLesson 821/887← PrevNext →
ArticlePublished 7 Aug 20262 min readBy Kevin Joginseparable algebraalgebra splittingidempotentBerlekamp
On this page

Ask about this page

KEVOS AISplitting Separable Algebras over Finite Fields

KEVOS knowledge first · trusted web sources when needed

Maximal Orders and Decomposition II

Splitting Separable Algebras over Finite Fields

Decomposing a finite-dimensional commutative algebra over a finite field into its simple components, generalising polynomial factorisation.

Engineering / MathematicsMaximal Orders and Decomposition II2 min readKV-MATH-0616

Splitting a commutative algebra over a finite field into fields is the generalisation of polynomial factorisation, and it is the engine of the Buchmann-Lenstra decomposition method.

The setting

A finite-dimensional commutative algebra over a finite field, with no nilpotent elements, decomposes as a direct product of finite fields. The task is to find that decomposition explicitly.

A = F_1 x F_2 x ... x F_gEach F_i a finite extension of the base field.

Key point

This is exactly Berlekamp's situation generalised. For the algebra of polynomials modulo a squarefree polynomial, the components correspond to the irreducible factors — see Berlekamp.

The Berlekamp subalgebra generalises

Elements fixed by Frobenius form a subalgebra whose dimension equals the number of components. It is computed as a kernel.

B = kernel of (Frobenius - identity) on ADimension equals the number of simple components.

Splitting a separable commutative algebra

  1. Remove nilpotentsQuotient by the radical if the algebra is not already reduced.
  2. Build the Frobenius matrixThe p-power map is linear over the prime field.
  3. Compute the kernelOf Frobenius minus the identity.
  4. Read the component countThe kernel dimension.
  5. SplitUse a non-trivial kernel element: its minimal polynomial factors, and the factors give idempotents separating the components.

Idempotents

The decomposition is equivalent to finding a complete set of orthogonal idempotents — one for each component. Each idempotent is obtained from a kernel element by evaluating a polynomial constructed from its minimal polynomial factors.

Key point

Idempotents are the practical output. Multiplying by an idempotent projects onto the corresponding component, which is how the individual prime ideals are extracted in Buchmann-Lenstra.

Large base fields

Pitfall

The naive splitting loop runs over elements of the base field and is therefore proportional to its size. For large base fields a probabilistic splitting step in the style of Cantor-Zassenhaus must be used instead.

Application to prime decomposition

The order modulo p, quotiented by its radical, is exactly such an algebra. Its simple components correspond to the prime ideals above p, and their degrees are the residue degrees. Splitting the algebra therefore decomposes the prime.

Note

This works regardless of whether p divides the index, which is precisely why the method handles the hard cases that simple decomposition cannot.

Source. Henri Cohen, A Course in Computational Algebraic Number Theory, Springer GTM 138 — 6.2.4. Structural reference unverified: the source file was not available during authoring; chapter and section numbers are taken from the published edition and have not been checked against a physical copy.

Related pages

  • The Berlekamp Factorisation Algorithm
  • Newton Polygon Methods for Prime Decomposition
  • The Buchmann-Lenstra Prime Decomposition Method

Continue learning

Newton Polygon Methods for Prime DecompositionGuide · Engineering MathematicsNEXT LESSON →The Buchmann-Lenstra Prime Decomposition MethodArticle · Engineering MathematicsRadical Computation and the Ring of MultipliersArticle · Engineering MathematicsThe Galois Group Computation ProblemGuide · Engineering Mathematics
KEVOS · Engineering, manufacturing and project improvement
ArticlesServicesCase studiesAboutContact
© 2026 KEVOS®