Let G be a group with subgroup H. ...Consider the relation ...x ∼ y ...⇔ ...y = h x , some h ∊ H. ...[ equivalently ...y x–1 ∊ H. ]
This is an equivalence relation: ...x = 1 x ; ...x = h–1 y ; ...z = g y ⇒ z = g ( h x ) = ( g h ) x .
For any x ∊ G , the right coset of x in H , ......H x ...:= ...[ x ] ...= ...{ y ∊ G | y = h x , some h ∊ H } , ...i.e. the equivalence class of x in the above relation.
Analagously, the left coset of x in H , ......x H ...:= ...{ z ∊ G | z = x h , some h ∊ H } ...( using the relation ...x ∼ z ...⇔ ...z = x h , some h ∊ H. )
In general ...x H ...≠ ...H x .
The set of right cosets in H is denoted ...H G ...:= ...{ H x | x ∊ G } , ...and the set of left cosets in H is denoted ...G ⁄ H ...:= ...{ x H | x ∊ G }
If G is finite, then the number of elements in a coset is the same as the number of elements in H , ...i.e. ...| H | .
For suppose H = { h1 , ⋯ , hn } , ...n distinct hi, ...then ...H x = { h1 x , ⋯ , hn x } . ...
Then for i ≠ j , ...hi x = hj x ...⇒ ...hi x x–1 = hj x x–1 ...⇒ ...hi = hj ...contradiction.
So the number of cosets in G is | G | ⁄ | H | . ...This number is known as the index of H in G , ...and is denoted ...| G : H | .
If G is an abelian group ( with group operation '+' ) then ...x ∼ y ...if ...x – y ∊ H , ...and the left and right cosets coincide, ...
i.e. ...x + H = H + x , ...∀ x . ...H G ...= ...G ⁄ H ...= ...{ x + H | x ∊ G}
Define the operation ...⊕ : ( G ⁄ H ) × ( G ⁄ H ) → ( G ⁄ H ) ...by ...( x + H ) ⊕ ( y + H ) ...= ...( x + y ) + H .
With this operation ...G ⁄ H ...is a group:
So if G is an abelian group with subgroup H , ...G ⁄ H ...with the operation ⊕ is called the quotient group of G by H .
Example: ...Consider the set of integers (positive, negative, and zero), Z, under addition. This is a group. Choose any integer ...n , ...then put ...H ...= ...nZ ...= ...{ n z | z ∊ Z } ...= ...{ 0, ±1n , ±2n , ⋯ } . nZ < Z , ...since n x – n y ...= ...n ( x – y ) .
The equivalence relation: ...x ∼ y ...⇔ ...( x – y ) ∊ nZ ...⇔ ...( x – y ) = n z , some z ∊ Z , ...we write ...x ≡ y ( mod n ) . ...The cosets defined by this relation are of the form ...x + nZ ...= ...{ x + n z | z ∊ Z } ...= ...{ x , x ± n , x ± 2n , ⋯ } .
Write ...Zn ...:= ...Z ⁄ nZ ...= ...{ x + nZ } . ...Note that ...[ wn ] ...= ...[ 0 ] , ...any w ∊ Z , ...so ...Zn ...= ...{ [ 0 ] , [ 1 ] , ⋯ , [ n – 1 ] } , ...which is written as ...{ 0 , 1 , ⋯ , n – 1 | n ≡ 0 }
Define ⊕ on Zn by: ......x + nZ ...⊕ ...y + nZ ...= ...x + y + nZ , ...then (Zn , ⊕ ) is isomorphic to the cyclic group of order n .
For illustrative purposes consider ...n = 3 . H ...= ...3Z ...= ...{ ⋯ –6, –3, 0, 3, 6, ⋯ }, ...x ∼ y ...⇔ ...x – y is divisible by 3, and we have
If we just write ...0 ...for ...0 + 3Z ...etc. then we have the "addition" table:
|
Addition table for (quotient) group ...Z ⁄ 3Z |
The mapping ...f : G → G ⁄ H ...given by ...f ( x ) ...= ...x + H , ...is called the canonical mapping .
f ...is a (surjective) homomorphism ...[ f ( x ) + f ( y ) ...= ...( x + H ) ⊕ ( y + H ) ...= ...( x + y ) + H ...= ...f ( x + y )] , ...so ...G ⁄ H ...is also abelian.
If G and H are abelian groups, and ...f : G → H ...is a homomorphism, then ...G ⁄ ker f ...≅ ...Im f ...(isomorphism) , ...where the isomorphism ...φ : G ⁄ ker f → Im f ...is given by ...φ ( x + ker f ) ...= ...f ( x ) .
|
φ is well-defined: ...x + ker f = x' + ker f ...⇒ ...x – x' ∊ ker f ...⇒ ...f ( x – x' ) = 0 ...⇒ ...f ( x ) = f ( x' ) ...so ...φ ( x + ker f ) = φ ( x' + ker f ) . φ is a homomorphism: ...φ ( ( x + ker f ) ⊕ ( y + ker f ) ) ...= ...φ ( ( x + y + ker f ) ) ...= ...f ( x + y ) ...= ...f ( x ) + f (y ) ...= ...φ ( x + ker f ) + φ ( y + ker f ) φ is injective: ...( x + ker f ) ≠ ( y + ker f ) ...⇒ ...x – y ∉ ker f ...⇒ ...f ( x – y ) ≠ 0 ...⇒ ...f ( x ) ≠ f ( y ) ...i.e. ...φ ( x + ker f ) ≠ φ ( y + ker f ) φ is surjective: ...y ∊ Im f ...⇒ ...y = f ( x ) = φ ( x + ker f ) , ...some x ∊ G . |
If G and H are any two groups, and ...f : G → H ...is a homomorphism then ......f injective ...⇔ ...ker f = { 1G }
f injective ...⇒ ...f ( x ) ≠ f ( 1G ) , if x ≠ 1G .
Conversely, if ...ker f = { 1G } , ...then ...x , y ∊ G , f ( x ) = f ( y ) ...⇔ ...f ( x y–1 ) = f ( x ) f ( y–1 ) = f ( x ) ( f ( y ) )–1 = 1H = f ( 1G )
...⇔ ...x y–1 = 1G ...⇔ ...x = y .