The Berlekamp-Massey algorithm is used for finding the minimal polynomial of a linearly recurrent sequence. The algorithm itself is presented at the end of this article.

To find the minimal polynomial, we need to be given an upper bound $m$ on its degree; having done so, the minimal polynomial is uniquely determined by the first $2m$ elements of $a$ (since we need to get $m$ equations to solve for the unknowns $c_{1},\ldots c_{m}$).

There is another way to determine the minimal polynomial, originally presented by Dornstetter, which uses the Euclidean Algorithm. It can be shown that the characteristic polynomial of a sequence is the unique monic polynomial $C(x)$ of least degree for which the infinite product

has finitely many nonzero terms. (In fact, the nonzero terms will have coefficients up to $x^{{k-1}}$ where $k$ is the degree of $C$).

We can rewrite this as

where $R(x)$ is a remainder polynomial of degree ¡ $m$, and $Q(x)$ is a quotient polynomial. Denote by $A(x)$ the sum $\Sigma_{{i=1}}^{{2m}}a_{i}x^{{i-1}}$.

This is where the Euclidean Algorithm comes in; if we take the GCD of $A(x)$ and $x^{{2m}}$, keeping track of remainders, we get two sequences $P_{i}(x),Q_{i}(x)$ such that

forms a series of polynomials whose degree is decreasing; as soon as this degree is less than $m$, we have the needed polynomials with $C=P_{i},Q=Q_{i}$.

There is more info about the Extended Euclidean Algorithm in “Modern Computer Algebra” by von zur Gathen and Gerhard.

(Berlekamp’s algorithm proper to come)

The original algorithm is from Algebraic Coding Theory by Elwyn R. Berlekamp, McGraw-Hill, 1968. Its application to linearly recurrent sequences was noted by J.L.Massey, in “Shift-register synthesis and BCH decoding”, 1969.