Lecture 6: Finite Fields (PART 3)

PART 3: Polynomial Arithmetic

Theoretical Underpinnings of AES and ECC

Lecture Notes on “Computer and Network Security”

by Avi Kak ([email protected])

April 26, 2009

c
2009 Avinash Kak, Purdue University

Goal of this part:

• To review polynomial arithmetic.

Goal for the upcoming part:

• To review finite fields of the form GF (2n)

1
 6.1: Polynomial Arithmetic

• A polynomial is an expression of the form

anxn + an−1 xn−1 + ...... + a1x + a0

for some non-negative integer n and where the coefficients a0,

a1, ...., an are drawn from some designated set S. S is called
the coefficient set.

• When an 6= 0, we have a polynomial of degree n.

• A zeroth-degree polynomial is called a constant polyno-

mial.

• Polynomial arithmetic deals with the addition, subtraction,

multiplication, and division of polynomials.

• Note that we have no interest in evaluating the value of

a polynomial for a specific value of the variable x.

2
 6.2: Arithmetic Operations on Polynomials

• We can add two polynomials:

f (x) = a2 x2 + a1 x + a0
g(x) = b1 x + b0
f (x) + g(x) = a2x2 + (a1 + b1)x + (a0 + b0)

• We can subtract two polynomials:

f (x) = a2 x2 + a1 x + a0
g(x) = b 3 x3 + b 0
f (x) − g(x) = −b3x3 + a2x2 + a1x + (a0 − b0)

• We can multiply two polynomials:

f (x) = a2 x2 + a1 x + a0
g(x) = b1 x + b0
f (x) × g(x) = a2b1x3 + (a2b0 + a1b1)x2 + (a1b0 + a0b1)x + a0b0

• We can divide two polynomials (result obtained by long division):

f (x) = a2 x2 + a1 x + a0
g(x) = b1 x + b0
f (x) / g(x) = ?

3
 6.3: Dividing One Polynomial by Another
Polynomial Using Long Division

• Let’s say we want to divide the polynomial 8x2 + 3x + 2 by

the polynomial 2x + 1:

• In this example, our dividend is 8x2 + 3x + 2 and the divisor

is 2x + 1. We now need to find the quotient.

• Long division for polynomials consists of the following steps:

– Arrange both the dividend and the divisor in the descending

powers of the variable.

– Divide the first term of the dividend by the first term of the
divisor and write the result as the first term of the quotient.
In our example, the first term of the dividend is 8x2 and the
first term of the divisor is 2x. So the first term of the quotient
is 4x.

– Multiply the divisor with the quotient term just obtained and
arrange the result under the dividend so that the same powers
of x match up. Subtract the expression just laid out from the
4
 dividend. In our example, 4x times 2x + 1 is 8x2 + 4x.
Subtracting this from the dividend yields −x + 2.

– Consider the result of the above subtraction as the new divi-

dend and go back to the first step. (The new dividend in our
case is −x + 2).

• In our example, dividing 8x2 + 3x + 2 by 2x + 1 yields a

quotient of 4x − 0.5 and a remainder of 2.5.

• Therefore, we can write

2.5
8x2 + 3x + 2 = 4x − 0.5 +
2x + 1

5
6.4: Arithmetic Operations on Polynomials Whose
Coefficients Belong to a Finite Field

• Let’s consider the set of all polynomials whose coefficients belong

to the finite field Z7 (which is the same as GF (7)). See Section
5.13 of Lecture 5 for the GF (p) notation.

• Here is an example of adding two such polynomials:

f (x) = 5x2 + 4x + 6
g(x) = 5x + 6
f (x) + g(x) = 5x2 + 2x + 5

• Here is an example of subtracting two such polynomials:

f (x) = 5x2 + 4x + 6
g(x) = 5x + 6
f (x) − g(x) = 5x2 + 6x

since the additive inverse of 5 in Z7 is 2 and that of 6 is 1. So

4x − 5x is the same as 4x + 2x and 6 − 6 is the same as
6 + 1, with both additions modulo 7.

6
• Here is an example of multiplying two such polynomials:
f (x) = 5x2 + 4x + 6
g(x) = 5x + 6
f (x) × g(x) = 4x3 + x2 + 1

• Here is an example of dividing two such polynomials:

f (x) = 5x2 + 4x + 6
g(x) = 2x + 1
f (x) / g(x) = 6x + 6

If you multiply the divisor 2x + 1 with the quotient 6x + 6,

you get the dividend 5x2 + 4x + 6.

7
 6.5: Dividing Polynomials Defined over a Finite
Field

• First note that we say that a polynomial is defined over a

field if all its coefficients are drawn from the field. It is also
common to use the phrase polynomial over a field to convey
the same meaning.

• Dividing polynomials defined over a finite field is a little bit more

frustrating than performing other arithmetic operations on such
polynomials. Now your mental gymnastics must include both
additive inverses and multiplicative inverses.

• Consider again the polynomials defined over GF (7).

• Let’s say we want to divide 5x2 + 4x + 6 by 2x + 1.

• In a long division, we must start by dividing 5x2 by 2x. This re-

quires that we divide 5 by 2 in GF (7). Dividing 5 by 2 is the same
as multiplying 5 by the multiplicative inverse of 2. Multiplicative
inverse of 2 is 4 since 2 × 4 mod 7 is 1. So we have
8
 5
= 5 × 2−1 = 5×4 = 20 mod 7 = 6
2
Therefore, the first term of the quotient is 6x. Since the product
of 6x and 2x + 1 is 5x2 + 6x, we need to subtract 5x2 + 6x from
the dividend 5x2 + 4x + 6. The result is (4 − 6)x + 6, which
(since the additive inverse of 6 is 1) is the same as (4 + 1)x + 6,
and that is the same as 5x + 6.

• Our new dividend for the next round of long division is therefore
5x + 6. To find the next quotient term, we need to divide 5x by
the first term of the divisor, that is by 2x. Reasoning as before,
we see that the next quotient term is again 6.

• The final result is that when the coefficients are drawn from the
set GF (7), 5x2 + 4x + 6 divided by 2x + 1 yields a quotient
of 6x + 6 and the remainder is zero.

• So we can say that as a polynomial defined over the field GF (7),

5x2 + 4x + 6 is a product of two factors, 2x + 1 and 6x + 6.
We can therefore write

5x2 + 4x + 6 = (2x + 1) × (6x + 6)

9
6.6: Let’s Now Consider Polynomials Defined over
GF (2)

• Recall from Section 5.13 of Lecture 5 that the notation GF (2)

means the same thing as Z2. We are obviously talking about
arithmetic modulo 2.

• First of all, GF (2) is a sweet little finite field. Recall that the
number 2 is the first prime. (A prime has exactly two distinct
divisors, 1 and itself.)

• GF (2) consists of the set {0, 1}. The two elements of this set
obey the following addition and multiplication rules:
0 + 0 = 0 0 X 0 = 0
0 + 1 = 1 0 X 1 = 0
1 + 0 = 1 1 X 0 = 0
1 + 1 = 0 1 X 1 = 1

0 - 0 = 0
1 - 0 = 1
0 - 1 = 0 + 1 = 1
1 - 1 = 1 + 1 = 0

10
• So the addition over GF (2) is equivalent to the logical XOR
operation, and multiplication to the logical AND operation.

• Examples of polynomials defined over GF (2):

x3 + x2 − 1
−x5 + x4 − x2 + 1
x + 1

11
6.7: Arithmetic Operations on Polynomials Over
GF (2)

• Here is an example of adding two such polynomials:

f (x) = x2 + x + 1
g(x) = x + 1
f (x) + g(x) = x2

• Here is an example of subtracting two such polynomials:

f (x) = x2 + x + 1
g(x) = x + 1
f (x) − g(x) = x2

• Here is an example of multiplying two such polynomials:

f (x) = x2 + x + 1
g(x) = x + 1
f (x) × g(x) = x3 + 1

12
• Here is an example of dividing two such polynomials:
f (x) = x2 + x + 1
g(x) = x + 1
1
f (x) / g(x) = x +
x + 1

If you multiply the divisor x + 1 with the quotient x, you get

x2 + x that when added to the remainder 1 gives us back the
dividend x2 + x + 1.

13
6.8: So What Sort of Questions does Polynomial
Arithmetic Address?

• Given two polynomials whose coefficients are derived from a set

S, what can we say about the coefficients of the polynomial that
results from an arithmetic operation on the two polynomials?

• If we insist that the polynomial coefficients all come from a par-

ticular S, then which arithmetic operations are permitted and
which prohibited?

• Let’s say that the coefficient set is a finite field F with its own
rules for addition, subtraction, multiplication, and division, and
let’s further say that when we carry out an arithmetic operation
on two polynomials, we subject the operations on the coefficients
to those that apply to the finite field F . Now what can be said
about the set of such polynomials?

14
 6.9: When is Polynomial Division Permitted?

• Polynomial division is obviously not allowed for polynomials that

are not defined over fields. For example, for polynomials defined
over the set of all integers, you cannot divide 4x2 + 5 by the
polynomial 5x. If you tried, the first term of the quotient would
be (4/5)x where the coefficient of x is not an integer.

• You can always divide polynomials defined over a field.

• In general, for polynomials defined over a field, the division of

a polynomial f (x) of degree m by another polynomial g(x) of
degree n ≤ m can be expressed by

f (x) r(x)
= q(x) +
g(x) g(x)

where q(x) is the quotient and r(x) the remainder.

• So we can write for any two polynomials defined over a field

f (x) = q(x)g(x) + r(x)

15
 assuming that the degree of f (x) is not less than that of g(x).

• When r(x) is zero, we say that g(x) divides f (x). This fact can
also be expressed by saying that g(x) is a divisor of f (x) and
by the notation g(x)|f (x).

16
6.10: Irreducible Polynomials, Prime Polynomials

• When g(x) divides f (x) without leaving a remainder, we say g(x)

is a factor of f (x).

• A polynomial f (x) over a field F is called irreducible if f (x)

cannot be expressed as a product of two polynomials, both over
F and both of degree lower than that of f (x).

• An irreducible polynomial is also referred to as a prime poly-

nomial.

17
6.11: Polynomials Over a Field Constitute a Ring

• The group operator is polynomial addition, with the addition of

the coefficients carried out as dictated by the field used for the
coefficients.

• The polynomial 0 is obviously the identity element with respect

to polynomial addition.

• Polynomial addition is associative and commutative.

• The set of all polynomials over a given field is closed under poly-
nomial addition.

• We can show that polynomial multiplication distributes over poly-

nomial addition.

• We can also show polynomial multiplication is associative.

18
• Therefore, the set of all polynomials over a field constitutes a
ring. Such a ring is also called the polynomial ring.

• Since polynomial multiplication is commutative, the set of poly-

nomials over a field is actually a commutative ring.

• In light of the constraints we have placed on what constitutes a

polynomial, it does not make sense to talk about multiplicative
inverses of polynomials in the set of all possible polynomials that
can be defined over a finite field. (Recall that our polynomials do
not contain negative powers of x.)

• Nevertheless, as you will see in the next lecture, it is possible for

a finite set of polynomials, whose coefficients are drawn from a
finite field, to constitute a finite field.

19
 Acknowledgement

Thanks go to Ian Calvert from the University of Birmingham for

catching a potentially misleading typographical error on page 11 of
the previous version of these notes.

20