Академический Документы
Профессиональный Документы
Культура Документы
Outline
Groups Fields Binary Field Arithmetic Construction of Galois Field Basic Properties of Galois Field Computations Using Galois Field Arithmetic Vector Spaces
2
Groups
Groups
Let G be a set () of elements (). A binary operation * on G is a rule that assign to each pair of elements a and b a uniquely defined third element c=a*b in G. Definition 2.1: A group is a set G with a binary operator * that satisfies the following conditions : Closure (): a, b G a b G Associative (): a, b, c G a * ( b * c ) = ( a * b ) * c Identity (): e G s.t. a G, a * e = e * a = a This element e is called an identity element of G. Inverse (): a G , a ' G, s.t. a * a ' = a '* a = e The element a is called an inverse of a.
4
Groups
Theorem 2.1: The identity element in a group G is unique. pf : Suppose it is not unique (e and e ): e = e*e = e*e = e (from definition) Theorem 2.2: The inverse of a group element is unique. pf : Suppose it is not unique (a and a ):
a = a*e (identity) (inverse) (associative) (inverse) (identity)
5
Groups
Remarks:
A group G is said to be commutative (abelian group) if a, b G, a*b = b*a Order: The number of elements in a group. We denote it |G|. Finite group: A group of finite order. Examples of commutative groups:
integers under + rational numbers under {0,1} under 2*2 real-valued matrix under + {0, 1, 2, ., m-1} under modulo-m addition is commutative (see example 2.2) {1, 2, 3.p-1} under modulo- p multiplicator (p is a prime) is also commutative (see example 2.3)
Groups
Example 2.1
Consider the set of two integers, G = {0, 1}. Let us define a binary operation, denoted by , on G as follows : 0 0 = 0 01 = 1 10 = 1 11 = 0 This binary operation is called modulo-2 addition. The set G = {0, 1} is a group under modulo-2 addition. It follows from the definition of modulo-2 addition that G is close under and 0 is the identity element. The inverse of 0 is itself and the inverse of 1 is also itself. It is easy to show that is associative. Thus, G together with is a commutative group.
7
Groups
Example 2.2
Let m be a positive integer. Consider the set of integer G = {0, 1, 2, , m-1}. Let + denote real addition. Define a binary operation on G as follows: For any integers i and j in G, i j = r , where r is the remainder resulting from dividing i + j by m. The remainder r is an integer between 0 and m-1 (Euclids division algorithm) and is therefore in G. Hence G is closed under the binary operation , called modulo-m addition. First we see that 0 is the identity element. For 0 < i < m, i and mi are both in G. Since i + (m i) = (m i) + i = m
8
Groups
It follows from the definition of modulo-m addition that i (m i) = (m i) i = 0 Therefore, i and m-i are inverses to each other with respect to . It is also clear that the inverse of 0 is itself. Since real addition is commutative, it follows from the definition of modulo-m addition that, for any integers i and j in G, i j = j i . Therefore modulo-m addition is commutative. Next we show that modulo-m addition is also associative. Let i, j, and k be three integers in G. Since real addition is associative, we have i + j + k = (i + j) + k = i + (j + k)
9
Groups
Dividing i + j + k by m, we obtain i + j + k = qm + r, where q and r are the quotient and the remainder, respectively, and 0 r < m . Now, dividing i + j by m, we have i + j = q1m + r1, with 0 r1 < m (2.1) Therefore, i j = r1. Dividing r1 + k by m, we obtain r1 + k = q2m + r2 with 0 r2 < m (2.2) Hence r1 k = r2 and (i j) k = r2. Combining (2.1) and (2.2), we have i + j + k = (q1+q2)m + r2, This implies that r2 is also the remainder when i + j + k is divided by m. Since the remainder resulting from dividing an integer by another integer is unique, we must have r2 = r.
10
Groups
As a result, we have (i j) k = r. Similarly, we can show that i ( j k ) = r. Therefore (i j) k = i ( j k ) and modulo-m addition is associative. This concludes our proof that the set G = {0, 1, 2, , m-1} is a group under modulo-m addition. We shall call this group an additive group.
11
Groups
Example 2.3: Let p be a prime (e.g. p =2, 3, 5, 7, 11,). Consider the set of integers, G = {0, 1, 2, , p-1}. Let denote real multiplication. Define a binary operation $ on G as follows: For i and j in G, i $ j = r, where r is remainder resulting from dividing i j by p. The set G = {0, 1, 2, , p-1} is a group under modulo-p multiplication. First we note that i j is not divisible by p. Hence 0 < r < p and r is an element in G. Therefore, the set G is closed under the binary operation $, referred to as modulo-p multiplication. We can easily check that modulo-p multiplication is commutative and associative. The identity element is 1. The only thing left to be proved is that every element in G has an inverse.
12
Groups
Let i be an element in G. Since p is a prime and i<p, i and p must be relatively prime (i.e. i and p dont have any common factor great than 1). It is well known that there exist two integers a and b such that a i + b p =1 (2.3) and i and p are relatively prime (Euclids theorem). Rearranging (2.3), we have a i = - b p + 1. (2.4) This says that when a i is divided by p, the remainder is 1. If 0 < a < p, a is in G and it follows from (2.4) and the definition of modulo-p multiplication that a $ i = i $ a = 1.
13
Groups
Therefore a is the inverse of i. However, if a is not in G, we divide a by p , a = q p + r. Since a and p are relatively prime, the remainder r cannot be 0 and it must be between 1 and p-1. Therefore r is in G. Now combining (2.4) and (2.5), we obtain r i = - (b + qi)p + 1. Therefore r $ i = i $ r = 1 and r is the inverse of i. Hence any element i in G has an inverse with respect to modulo-p multiplication. The group G = {0, 1, 2, , p-1} under modulo-p multiplication is called a multiplicative group.
Subgroup
Theorem 2.3: Let G be a group under the binary operation *. Let H be a nonempty subset of G. Then H is a subgroup of G if the following conditions hold: (i) H is closed under the binary operation *. (ii) For any element a in H, the inverse of a is also in H. Proof: Condition (ii) says that every element of H has an inverse in H. Condition (i) & (ii) ensure that the identity element of G is also in H. (a*a=e is an element of H) Because the elements in H are elements in G, the associative condition on * holds automatically. H satisfies all the conditions of a group.
15
Coset
Definition 2.2: Let H be a subgroup of a group G with binary operation *. Let a be an element of G. Then the set of elements a * H {a * h : h H } is called a left coset of H; the set of elements H * a {h * a : h H } is called a right coset of H. If the group G is commutative, then every left coset is identical to every right coset. Example: Consider the additive group G={1,2,3,,15} under modulo-16 addition. Subgroup H={0,4,8,12}. The coset 3 H 3 H = {3 0,3 4,3 8,3 12}
= {3, 7,11,15} 7 H = {7 0, 7 4, 7 8, 7 12} = {7,11,15,3}
16
Coset
There are only four distinct cosets of H:
Coset
Suppose two elements, say a*h and a*h, in a*H are identical, where h and h are two distinct elements in H. Let a-1 denote the inverse of a with respect to the binary operation *. Then a-1 *(a*h)= a-1 *(a*h), (a-1 *a)*h=(a-1 *a)*h, e *h=e *h, h=h. This result is a contradiction to the fact that all the elements of H are distinct. Therefore, no two elements in a coset are identical.
18
Coset
2.5: No two elements in two different cosets of a subgroup H of a group G are identical. Proof: Let a*H and b*H be two distinct cosets of H, with a and b in G. Let a*h and b*h be two elements in a*H and b*H, respectively. Suppose a*h=b*h. Let h-1 be the inverse of h. (a*h)* h-1 =(b*h)* h-1 a*(h* h-1) =b*(h* h-1) a*e=b*h where (h= h* h-1) is an element in H.
19
Coset
The equality a=b*h implies that a * H = ( b * h '') * H
= {( b * h '') * h : h H } = {b * ( h ''* h ) : h H } = {b * h ''' : h ''' H } = b*H This result says that a*H and b*H are identical, which is a contradiction to the given condition that a*H and b*H are two distinct cosets of H. Therefore, no two elements in two distinct cosets of H are identical.
20
Coset
From Theorem 2.5 and 2.5, we obtain the following properties of cosets of a subgroup H of a group G: Every element in G appears in one and only one coset of H; All the distinct cosets of H are disjoint; The union of all the distinct cosets of H forms the group G. All the distinct cosets of a subgroup H of a group G form a partition of G, denoted by G/H.
21
Lagranges Theorem
Lagranges Theorem: Let G be a group of order n, and let H be a subgroup of order m. Then m divides n, and the partition G/H consists of n/m cosets of H. Proof: Every coset consists of m elements of G. Let i be the number of distinct cosets of H. Since n=im, m divides n and i=n/m.
22
Fields
Fields
Roughly speaking, a field is a set of elements in which we can do addition, subtraction, multiplication, and division without leaving the set. Addition and multiplication must satisfy the commutative, associative, and distributive laws. Definition 2.3: Let F be a set of elements on which two binary operations, called addition + and multiplication , are defined. The set F together with the two binary operations + and is a field if the following conditions are satisfied: F is a commutative group under addition +. The identity element with respect to addition is called the zero element or the additive identity of F and is denoted by 0.
24
Fields
The set of nonzero elements in F is a commutative group under multiplication . The identity element with respect to multiplication is called the unit element or the multiplicative identity of F and is denoted by 1. Multiplication is distributive over addition; that is, for any three elements a, b, and c in F, a (b+c)=a b+a c
25
Fields
A field consists of at least two elements, the additive identity and the multiplicative identity. The number of elements in a field is called the order of the field. A field with finite number of elements is called a finite field. In a field, the additive inverse of an element a is denoted by a and the multiplicative inverse of a is denoted by a-1 provided that a0. Subtracting a field element b from another field element a is defined as adding the additive inverse b of b to a. [a-ba+(-b)]. If b is a nonzero element, dividing a by b is defined as multiplying a by the multiplicative inverse b-1 of b. [a/ba b-1].
26
Fields
Property I. For every element a in a field, a 0=0 a=0.
Proof. a=a1=a (1+0)=a+a 0 Adding a to both sides of the equality above, we have: -a+a= -a+a+a 0 0=0+a 0 0=a 0 Similarly, we can show that 0 a=0. Therefore, we obtain a 0=0 a=0.
Property II. For any two nonzero elements a and b in a field, a b0.
Proof. From definition, nonzero elements of a field are closed under multiplication.
27
Fields
Property III. ab=0 and a0 imply that b=0.
This is a direct consequence of Property II.
Fields
Some examples:
R (real number set) C (complex number) Q (Rational number) GF(q) exists if q = p m, p is a prime Ex. Binary field GF(2) with modulo-2 addition
+ 0 1 0 0 1 0 0 0
infinite fields
1 1 0 1 0 1
modulo-2 multiplication
29
0 1
Fields
Ex: GF(p) , p is a prime. (Prime Field)
{0, 1, 2, .p-1} is an abelian group under modulo-p addition. {1, 2, p-1} is an abeian group under modulo-p multiplication. Fact: real number multiplication is distributive over real number addition. This implies that modulo-p multiplication is distributive over modulo-p addition. {0, 1, 2, .p-1} is a field of order p under modulo-p addition and multiplication. In fact, for any positive integer m, it is possible to extend the prime field GF(p) to a field of pm elements called an extension field of GF(p) and is denoted by GF(pm). Furthermore, the order of any finite field is a power of a prime. Finite fields are also called Galois field.
30
Fields
Def: Characteristic of GF(q)
Consider the unit element 1 in GF(q); a smallest positive integer s.t. 1 = 0 , then is called the characteristic of GF(q)
EX.
i =1
1 = 0
i =1
31
Fields
Theorem 2.7: The characteristic of GF(q) is prime.
(pf)
k, m < k m km 1. 1 = 1 = 1 = 0 1 1 1 1 k m 1 = 0 or 1 = 0
= km
1 1
i =1
(otherwise
mk i =1
i =1
1 = 0
32
m k < )
Fields
Remarks:
1, 1+1, 1+1+1, , 1+1++1=0 are distinct elements in GF(q), which form a subfield GF( ) of GF(q) If q , then q is a power of (proven in later)
n Let a GF (q), a 0, a smallest positive integer n s.t. a = 1 . n is called the order of the field element.
33
Fields
Remarks:
a, a 2=a a, a 3=a a a, , a n 1 a n =1 are all distinct, which form , a group under the multiplication of GF(q). (pf): Closure if i + j n, a i a j = a i + j
if i + j > n, we have i + j = n + r
where 0 r n ai a j = ai+ j = a n a r = a r i Inverse For 1 i < n, a n i is the multiplicative inverse of a Since the powers of a are nonzero elements in GF(q), they satisfy the associative and communitative laws.
34
Fields
Def: (Cyclic)
A Group is said to be cyclic, if there exists an element in the group whose powers constitute the whole group.
Fields
Def: (Primitive)
a GF ( q )& a 0 , a is said to be primitive if the order of a is q-1
Remarks:
The powers of a primitive element generate all the nonzero elements of GF(q) Every finite field has a primitive element. Example. GF(5) 2' = 2, 2 2 = 4, 2 3 = 3, 2 4 = 1, 25 = 2 2 is a primitive elements. primitive elements are useful for constructing fields. Example. GF(7) 3 is a primitive element, the order of 4 is 3, which divides 6.
36
Binary Arithmetic
In general, we can construct codes with symbols from any Galois field GF(q), where q is either a prime p or a power of p; however, codes with symbols from the binary field GF(2) or its extension GF(2m) are most widely used in digital data transmission and storage systems. In this text, we are concerned only with binary codes and codes with symbol from the field GF(2m). Most of the results presented in this text can be generalized to codes with symbols from any finite field GF(q) with q=2 or 2m. In binary arithmetic, we use modulo-2 addition and multiplication, which are defined by Tables 2.3 and 2.4, respectively.
38
1 0 1
0 1 = 1 1
0 1
1 1
1 1
1 1
+ 0
1 1
0 1
= 1 1 1 0 + 0 1 = 1
1 1 0 1 0 1 1 1 1 1 y= = =1 1
39
1 1 1 1 0 0 1 1 1 0 z= = =0 1
Def: f ( x ) = f + f x + ... + f x
0 1 n
fi GF(2)
Remarks:
Polynomials over GF(2) with degree = 1 ex : x , 1+x Polynomials over GF(2) with degree = 2 ex : x 2 , 1 + x 2 , x + x 2 , 1 + x + x 2 In general, with degree = n we have 2 n polynomials.
40
f ( x) + g ( x) = g ( x) + f ( x) f ( x) g ( x) = g ( x) f ( x)
41
Def: (Irreducible) p(x) GF(2) [x] with deg[p(x)]=m is said to be irreducible over
GF(2) if p(x) is not divisible by any polynomial over GF(2) of degree less than m but greater than zero.
43
is an irreducible polynomial with degree 2. x 3 + x + 1 is also an irreducible poly with degree 3. In general, for any m1 , there exists an irreducible polynomial of degree m.
+ 1. degree m divides x e.g. 3 23 1 ( x + x + 1) | x +1= x7 +1
44
Def: (Primitive)
Remarks:
For a given m>0, there may be more than one primitive polynomials of degree n. Lists of primitive polynomials (see p.42 Table 2.7)
45
46
00 = 0 0 1 = 1 0 = 0 1 1 = 1 0 = 0 = 0 1 = 1 =
2 = , 3 = j = ,
j times
48
i j = j i = i + j
F = { 0,1, ,... j ,...} with 1 be sometime denoted by 0
49
2m 1
+ 1 = q ( ) p ( ) = q( ) 0 = 0
2 m 1
=1
2m 2
50
p ( x ) x i (1 + x j i ) p ( x ) x j i + 1
i = qi ( ) 0 + ai ( )
i=0 distinct nonzero poly. of over GF (2) with degree m-1 or less.
53
{ }
2m 2
the power representation (for multiplication) the polynomial representation (for addition)
Set p( ) = 1 + +
= 0
2
=1+
2 3
5 6 7
= = =
4 5 6
= (1 + ) = + 2 = ( + = (
2
) = =
+ +
+ 3) =
3
+1+
3
=1+ +
8 9 10 11 12 13 14
=1+
2 3 2 3
= + = + =1+ =1+
=1+ +
2
(1110) (0111)
3
+
2
=1+ +
2 3
+
3
57
a 7 = a 12 ,
12
=
7
19
15
= 1)
12
12
= =
7
15 12
=
22
12
10
= ( + 2 ) + (1 + + 3 ) = 1 + 2 + 3 = 13
1 + 5 + 10 = 1 + ( + 2 ) + (1 + + 2 ) = 0
Another useful representation of field elements in GF(2m ) : a0 + a1 + ... + am 1 m 1 m tuple (a0 , a1 ,......am1 )
58
{0,1, , } {0,1, , + 1}
+ 0 1 0 0 1 1 1 0
2 +1 1+
1 0
0 0 0
1 0 1
+1
0
0 1
+1 0 1 2 +1
+1
+1 0 +1 1 0 +1 1
59
61
(pf):
[ f ( x )]2 = f ( x 2 ) f ( 2 ) = [ f ( )]2 = 0 2 = 0
63
2m 1
+1
(pf):
2m 1
2 m 1
Remarks:
is a root of x
2m
+x
65
1 ( ) = 0 or 2 ( ) = 0
(x) is not a minimal poly.
of . If f ( ) = 0, then (x) f ( x)
69
f ( x) = ( x + )
i =0
e 1
2i
is an irreducible poly over GF(2) (pf): prove that f ( x ) GF (2) [x ] by first prove [ f ( x)]2 = f ( x 2 ) [ f ( x )] = ( x + 2 i = 0
2
e 1
e 1 2 i =0 e 1
i
e 1 2i 2 ) = ( x + ) i =0
2i 2i +1
= (x + ( + )x +
= (x +
2 i =0 2i +1
2i
70
= (x + )(x + )
2 2 i=1
e1
i =1
2i
2e
= (x + )
2 i =0
e 1
2i
2e
[ f ( x) ]
e i =0
= ( f 0 + f1 x + + f e x e ) 2
2i e e i+ j i =0 j =0 i j
= f i x + (1 + 1) fi f j x
2
= fi x
2 i =0
2i
71
This holds only when fi = 0 or 1 f ( x) GF (2)[x ] prove that f(x) is irreducible over GF(2)
if not, f (x ) = a (x )b (x )&f ( ) = a ( )b ( ) = 0 a ( ) = 0 or b ( ) = 0, if a ( ) = 0, a ( x ) has 2 2e 1 roots , , . (Theorem 2.11)
( x) = ( x + )
2i
e 1
i =0
= x4 + x3 + x2 + x + 1
73
r = 7 in GF(24 )
28
= , r
13
23
= 56 = 11
Hence ( x ) = a 0 + a1 x + a 2 x 2 + a3 x 3 + x 4 ( r ) = a 0 + a1r + a 2 r 2 + a3 r 3 + r 4 = 0
2 3 4 Using poly. representation for r, r ,r ,r
a0 + a1(1+ + 3 ) + a2 (1+ 3 ) + a3 (2 + 3 ) + 1+ 2 + 3 = 0
a 0 + a1 + a 2 + 1 = 0 a1=0 a3 +1 = 0 a1 + a 2 + a 3 + 1 = 0
74
(x) = 1+ x3 + x4
All the minimal poly. of elements in GF(24). (See p.52 Table2.9) Thm2.19: Let (x) be the minimal poly. of GF ( 2 m ) & deg[ ( x )] = e . Then e be the smallest integer s.t. 2 = . Moreover e m (direct from Thm2.18&2.19)
Remarks:
m f(x) be the minimal poly. of GF (2 ) with deg[ f(x)] = e then e | m (proof is omitted)
75
In the construction of the Galois field GF(2m), we use a primitive polynomial p(X) of degree m and require that the element be a root of p(X). Because the powers of generate all the nonzero elements of GF(2m), is a primitive element. In fact, all the conjugates of are primitive elements of GF(2m).
76
is a primitive element of GF (2 m )
its order is 2 m 1
For n 2 = 1, (2 m - 1) n 2
4 = 28 = 13
14 = 98 = 8 , 15 = 105 = 1
Clearly, = 7 is a primitive element of GF(24) 2 14 22 13 22 = , = , = 11 are all primitive elements of GF(24)
78
79
EX1:
2 7 4 8 10 + 11 14 1+ 3 X = = = = = 8 19 2 5 7 + + 1 1 2 8
1 2 12 4 1 7 12 8
81
Y =
4 + 16 + 3 9 = 8 = = 5 = 4 + 19 + 2
f ( 6 ) = 12 + 13 + = 0
f ( 10 ) = 20 + 17 + = 0
f ( x) = ( x + 6 )(x + 10 ) = 0
x = ,
6
10
82
Vector Spaces
Vector Spaces
Def: V be a set of elements with a binary operation + is defined.
F be a field. A multiplication operator between a F and v V is also defined. The V is called a vector space over the field F if: V is a commutative group under +
a F & v V a v V distributive law u ,v V and a,b F a (u + v ) = a u + a v (a + b) v = a v + b v
Associative Law ( a b ) v = a ( b v) 1v = v .
84
Vector Spaces
The elements of V are called vectors and the elements of the field F are called scalars. The addition on V is called a vector addition and the multiplication that combines a scalar in F and a vector in V is referred to as scalar multiplication (or product) The additive identity of V is denoted by 0.
85
Vector Spaces
Property III. For any scalar c in F and any vector v in V, (-c) v = c (- v ) = -(c v)
i.e., (-c) v or c (- v ) is the additive inverse of the vector c v. (Left as an exercise) Consider an ordered sequence of n components, ( a0 , a1 ,..., a n 1 ) , where each component ai is an element from the binary field GF(2) (i.e., ai = 0 or 1). This sequence is called an n-tuple over GF(2). Since there are two choices for each ai, we can construct 2n distinct n-tuples. Let Vn denote this set. Now we define an addition + on Vn as following : For any u = (u 0 , u1 ,..., u n 1 ) and v = (v0 , v1 ,..., vn 1 ) in V n , u + v = (u0 + v0 , u1 + v1 ,..., un1 + vn1 ) (2.7)
86
Vector Spaces
where ui+vi is carried out in modulo-2 addition. Clearly, u + v is also an n-tuple over GF(2). Hence Vn is closed under the addition. We can readily verify that Vn is a commutative group under the addition defined by (2.27) . (1) we see that allzero n-tuple 0 = (0, 0, ,0) is the additive identity. For any v in Vn , v + v = (v0 + v0 , v1 + v1 ,..., vn1 + vn1 ) = (0, 0, ,0) = 0 Hence, the additive inverse of each n-tuples in Vn is itself. Since modulo-2 addition is commutative and associative, the addition is also commutative and associative. Therefore, Vn is a commutative group under the addition. (2) we defined scalar multiplication of an n-tuple v in Vn
87
Vector Spaces
by an element a from GF(2) as follows : a (v0 , v1 ,..., vn 1 ) = ( a v 0 , a v1 ,..., a v n 1 ) (2.28) where avi is carried out in modulo-2 multiplication. Clearly, a (v0 , v1 ,..., vn 1 ) is also an n-tuple in Vn . If a = 1, 1 (v0 , v1 ,..., vn 1 ) = (1 v 0 ,1 v1 ,...,1 v n 1 ) = (v0 , v1 ,..., vn 1 ) By (2.27) and (2.28), the set Vn of all n-tuples over GF(2) forms a vector space over GF(2)
88
Vector Spaces
Example
Let n=2. The vector space V2 of all 2-tuples over GF(2) consists of the following 4 vectors : (0 0) (0 1) (1 0) (1 1) The vector sum of (0 0) and (0 1) is (0 0) + (0 1) = (0 + 0 0+1) = (0 1) Using the rule of scalar multiplication defined by (2.28), we get 0 (1 0) = (01 00) = (0 0) 1 (1 1) = (11 11) = (1 1)
V being a vector space of all n-tuples over any field F, it may happen that a subset S of V is also a vector space over F. Such a subset is called a subspace of V.
89
Vector Spaces
Theorem 2.18
Let S be a nonempty subset of a vector space V over a field F. Then S is a subspace of V if the following conditions are satisfied : (1) For any two vectors u and v in S, u + v is also a vector in S. (2) For an element a in F and any vector u in S, a u is also in S. (pf). Conditions (1) and (2) say simply that S is closed under vector addition and scalar multiplication of V. Condition (2) ensures that, for any vector v in S, its additive inverse (-1) v is also in S. Then, v + (-1)v = 0 is also in S. Therefore, S is a subgroup of V. Since the vectors of S are also vectors of V, the associative and distributive laws must hold for S. Hence, S is a vector space over F and is a subspace of V.
90
Vector Spaces
Let v1, v2, ,vk be k vectors in a vector space V over a field F. Let a1, a2, , ak be k scalars from F. The sum a1v1 + a2v2 + + akvk is called a linear combination of v1, v2, ,vk . Clearly, the sum of two linear combinations of v1, v2, ,vk , (a1v1 + a2v2 + + akvk ) + (b1v1 + b2v2 + + bkvk ) = (a1+b1 )v1 + (a2+b2 )v2 ++ (ak+bk )vk is also a linear combination of v1, v2, ,vk , and the product of a scalar c in F and a linear combination of v1, v2, ,vk , c(a1v1 + a2v2 + + akvk ) = (c a1 ) v1 + (c a 2 ) v 2 + ... + (c a k ) v k is also a linear combination of v1, v2, ,vk
Theorem 2.19
Let v1, v2, ,vk be k vectors in a vector space V over a field F. The set of all linear combinations of v1, v2, ,vk forms a subspace of V.
91
Vector Spaces
A set of vectors v1, v2, ,vk in a vector space V over a field F is said to be linearly dependent if and only if there exit k scalars a1, a2, , ak from F, not all zeros, such that a1v1 + a2v2 + + akvk = 0 A set of vectors v1, v2, ,vk is said to be linearly independent if it is not linearly dependent. That is, if v1, v2, ,vk are linearly independent, then a1v1 + a2v2 + + akvk 0 unless a1 = a2 = = ak = 0.
Vector Spaces
However, (1 0 1 1 0), (0 1 0 0 1), and (1 1 1 1 1) are linearly independent. A set of vectors is said to span a vector space V if every vector in V is a linear combination of the vectors in the set. In any vector space or subspace there exits at least one set B of linearly independent vectors which span the space. This set is called a basis (or base) of the vector space. The number of vectors in a basis of a vector space is called the dimension of the vector space. (Note that the number of vectors in any two bases are the same.)
93
Vector Spaces
Consider the vector space Vn of all n-tuples over GF(2). Let us form the following n n-tuples :
e 0 = (1 0 0 0 ... 0 0) e 1 = ( 0 1 0 0 ... 0 0)
e n- 1 = ( 0 0 0 0 ... 0 1),
where the n-tuple ei has only nonzero component at ith position. Then every n-tuple ( a0 , a1 ,..., a n 1 ) in Vn can be expressed as a linear combination of e0, e1,,en-1 as follows :
Vector Spaces
Therefore, e0, e1,,en-1 span the vector space Vn of all n-tuples over GF(2). We also see that e0, e1,,en-1 are linearly independent. Let u = (u 0 , u1 ,..., u n 1 ) and v = (v0 , v1 ,..., vn 1 ) be two n-tuples in Vn . We define the inner product (or dot product) of u and v as where uivi and uivi + ui+1vi+1 are carried out in modulo-2 multiplication and addition. Hence the inner product uv is a scalar in GF(2). If uv = 0, u and v are said to be orthogonal to each other. The inner product has the following properties :
uv = vu u(v+w) = uv + uw (au)v = a(uv)
95
Vector Spaces
Let S be a k-dimension subspace of Vn and let Sd be the set of vectors in Vn such that, for any u in S and v in Sd, uv = 0. The set Sd contains at least the all-zero n-tuple 0 = (0, 0, , 0), since for any u in S, 0u = 0. Thus, Sd is nonempty. For any element a in GF(2) and any v in Sd, 0 if a = 0 av ={ v if a = 1 Therefore, av is also in Sd . Let v and w be any two vectors in Sd . For any vector u in S, u(v+w) = uv + uw = 0 + 0 = 0. This says that if v and w are orthogonal to u, the vector sum v + w is also orthogonal to u. Consequently, v + w is a vector in Sd. It follows from Theorem 2.18 that Sd is also a subspace of Vn . This subspace is called the null (or dual) space of S. Conversely, S is also the null space of Sd.
96
Vector Spaces
Theorem 2.20
Let S be a k-dimension subspace of the vector space Vn of all n-tuples over GF(2). The dimension of its null space Sd is n-k. In other words, dim(S) + dim(Sd)= n.
97