Exercise 11.2.12

Let f 𝔽 p [ x ] be monic and separable of degree n > 1 , and assume that T 1 R has rank n 1 . By theorem 11.2.9, f is reducible. In this exercise, you will use the kernel of T 1 R to produce a nontrivial factorization of f .

(a)
Show that the constant polynomials in 𝔽 p [ x ] give a one-dimensional subset of the kernel of T 1 R .
(b)
Prove that there is a nonconstant polynomial h of degree < n such that f h p h . Parts (c),(d) and (e) will use h to produce a nontrivial factorization of f .
(c)
Explain why h p h = a 𝔽 p ( h a ) .
(d)
Use parts (b) and (c) to show that f = a 𝔽 p gcd ( f , h a ) .
(e)
Use deg ( h ) < n to show that f gcd ( f , h a ) . Conclude that the factorization of part (d) is nontrivial, i.e., gcd ( f , h a ) is a nonconstant factor of f of degree < n for at least two a 𝔽 p .

The basic idea of Berlekamp’s algorithm is that one can factor f into irreducibles by taking the gcd’s of the nontrivial factors gcd ( f , h a ) produced by part (e) as we vary h and a .

Answers

Proof.

(a)
Let α = [ i ] + f be the coset of the constant polynomial [ i ] 𝔽 p . Then ( T 1 R ) ( α ) = α p α = ( [ i ] p [ i ] ) + f = [ 0 ] + f = 0 ¯ , so α ker ( T 1 R ) . vect 𝔽 p ( [ 1 ] + f ) = { α R | [ i ] 𝔽 p , α = [ i ] + f }

is a one-dimensional subspace of the kernel of T 1 R , corresponding to the constant polynomials in 𝔽 p [ x ] .

(b)
By part (a), the rank of T 1 R is not n , and by hypothesis this rank is not n 1 , so the rank of T 1 R is less than n 1 , so the kernel of T 1 R has dimension at least 2.

Therefore there exists a polynomial h , with deg ( h ) < n , such that h ¯ = h + f ker ( T 1 R ) which is not in the one-dimensional subspace of part (a), thus h is not a constant polynomial. Since h ker ( T 1 R ) , h p + f = h + f , thus f h p h .

Conclusion: there is a nonconstant polynomial h of degree < n such that f h p h .

(c)
In k [ x ] , we know that x p x = a 𝔽 p ( x a ) .

The formal composition with h ( x ) gives

h p h = a 𝔽 p ( h a ) .

(d)
We know that if f 1 , , f s are polynomials in 𝔽 p [ x ] such that gcd ( f i , f j ) = 1 if i j , then gcd ( f , f 1 f s ) = gcd ( f , f 1 ) gcd ( f , f s ) .

Take f a = h a 𝔽 p [ x ] . Then a b gcd ( f a , f b ) = 1 , since

1 b a ( h a ) 1 b a ( h b ) = 1 .

Therefore, since f h p h ,

f = gcd ( f , h p h ) = gcd ( f , a 𝔽 p ( h a ) ) = a 𝔽 p gcd ( f , h a )
(e)
Since deg ( h ) < n , deg ( h a ) < n , therefore deg ( gcd ( f , h a ) ) < n . Moreover, since f 0 , deg ( gcd ( f , h a ) ) 0 , so f cannot divide gcd ( f , h a ) when a 𝔽 p .

Therefore the product a 𝔽 p gcd ( f , h a ) has more than one nonconstant factor (if all factors except one are constant, then f gcd ( f , h a ) for some a 𝔽 p , which is impossible).

So gcd ( f , h a ) is a nonconstant factor of f (and non associate to f ) for at least two a 𝔽 p .

User profile picture
2022-07-19 00:00
Comments